PDA

View Full Version : Baseball Prospectus


Ava83
06-09-2008, 10:06 PM
Would anyone know exactly (paraphrasing will work) how they come up with the following? I understand its a computer program that measures some stats and that is it.

http://www.baseballprospectus.com/statistics/ps_oddspec.php

Or do they use a magic 8 ball type thing with % instead?

anewman35
06-09-2008, 10:10 PM
Would anyone know exactly (paraphrasing will work) how they come up with the following? I understand its a computer program that measures some stats and that is it.

http://www.baseballprospectus.com/statistics/ps_oddspec.php

Or do they use a magic 8 ball type thing with % instead?

Even better looking (for us) version at http://www.baseballprospectus.com/statistics/ps_odds.php. They sort of explain it at the bottom - basically they put in a whole bunch of stats and do tons of simulations. Another site that does basically the same thing is www.coolstandings.com (http://www.coolstandings.com). It doesn't mean much, but it's fun to see it go up as we keep winning.

And, yes, somebody is going to complain that in 2005, it said we had an 88% chance of winning the division when we'd already cliniched. That's not really true, though - at the time, we'd only clinched a tie (which we would have won on tiebreaker) - they were saying we had an 88% percent chance of winning outright.

Ava83
06-09-2008, 10:18 PM
Even better looking (for us) version at http://www.baseballprospectus.com/st...cs/ps_odds.php (http://www.baseballprospectus.com/statistics/ps_odds.php). They sort of explain it at the bottom - basically they put in a whole bunch of stats and do tons of simulations. Another site that does basically the same thing is www.coolstandings.com (http://www.coolstandings.com/). It doesn't mean much, but it's fun to see it go up as we keep winning.



I honestly do not put too much stock in it...and I did read that bottom part but they lost me at regression...I was actually looking for a layman explanation. And I just read the log5 method...Algebra looked to be involved and I decided to stop reading :rolleyes:

Thank you so much though!

HomeFish
06-09-2008, 10:29 PM
The Angels only have a 40% chance of winning the AL West, while the Athletics have a 54% chance. The Angels are currently 4.5 games up.

It's funny, because in every other division the current leader is given an 80% chance to win.

Huisj
06-09-2008, 10:46 PM
Even better looking (for us) version at http://www.baseballprospectus.com/statistics/ps_odds.php. They sort of explain it at the bottom - basically they put in a whole bunch of stats and do tons of simulations. Another site that does basically the same thing is www.coolstandings.com (http://www.coolstandings.com). It doesn't mean much, but it's fun to see it go up as we keep winning.

And, yes, somebody is going to complain that in 2005, it said we had an 88% chance of winning the division when we'd already cliniched. That's not really true, though - at the time, we'd only clinched a tie (which we would have won on tiebreaker) - they were saying we had an 88% percent chance of winning outright.

I believe the situation was that the Sox had clinched a tie, and they had won the tie breaker based on head-to-head with Cleveland, so no matter what happened, they would be division champs, but BP said they didn't like the tie breaker rule, so it wasn't a part of their model, and they acted like they were proud of their model being wrong because they thought that was how it should have been.

DumpJerry
06-09-2008, 10:46 PM
They have a monkey throw darts at numbers which are scattered at random on a wall to figure out how many games each team will win.



The rest just falls in place.

Daver
06-09-2008, 10:48 PM
BP only publishes things that fit their theories.

Eddo144
06-09-2008, 10:57 PM
BP only publishes things that fit their theories.
They keep this page updated all year, regardless of which teams are actually in first place. :scratch:

I never get the pure hatred aimed at BP. For the most part, they're not too arrogant, despite how some true believers like the writers at FireJoeMorgan behave. It's not like they just pick some numbers to use - they run historical seasons to see how their projections match up with actual results. Just because they were skeptical of the early-season success the White Sox in 2005 doesn't make them out to get us.

(For those of you that are interested in some of BP's research, I recommend Baseball Between the Numbers, which is divided into 27 chapters that discuss everything from team building to how winning affects team revenue to the construction of an ideal batting order. In fact, in chapter 8.3 ("Is Wayne Huizenga a Genius?", which deals with the way all of the World Series winners since 1995 were built), they're even complementary to Kenny Williams. Shocking, I know. When discussing the 2005 White Sox, they write, "This was arguably the best job by a general manager heading into a championship year of any on this list" (p. 321).)

jabrch
06-10-2008, 12:04 AM
It is all about computer generated simulations. It means.... Yes.... NOTHING

FedEx227
06-10-2008, 02:45 AM
(For those of you that are interested in some of BP's research, I recommend Baseball Between the Numbers, which is divided into 27 chapters that discuss everything from team building to how winning affects team revenue to the construction of an ideal batting order. In fact, in chapter 8.3 ("Is Wayne Huizenga a Genius?", which deals with the way all of the World Series winners since 1995 were built), they're even complementary to Kenny Williams. Shocking, I know. When discussing the 2005 White Sox, they write, "This was arguably the best job by a general manager heading into a championship year of any on this list" (p. 321).)

Definitely a good read, there are some chapters where the BP arrogance comes in, but a lot of it is just good solid baseball research with plenty of graphs, etc.

Ava83
06-10-2008, 09:07 AM
Thanks All! I am reading Moneyball at the moment (with a grain of salt...but and interesting read) So this sparked my interest. I may pick up the other book mentioned next.

cws05champ
06-10-2008, 09:08 AM
What was their prediction for the Sox at the begining of the season?

kjhanson
06-10-2008, 09:47 AM
Hah....I think it was like 72-90 or 77-85. I'm going to look it up now though...

EDIT: It was 77-85. We may surpass that by 20 games now.

Lip Man 1
06-10-2008, 01:04 PM
And if that happens you can be absolutely sure that you WON'T be reading a story by the stat-geeks at BP on how they got it wrong (again...)

Lip

Eddo144
06-10-2008, 01:21 PM
For all those who argue that BP and sabermetrics are trying to make the game only about numbers or throw out the line "baseball isn't played on a computer," here's a nice article (http://sportsillustrated.cnn.com/2008/baseball/mlb/05/30/bp.500HRs/index.html?eref=T1) written by Joe Sheehan (of BP) for Sports Illustrated. Sheehan writes how you shouldn't celebrate numbers, like 500 and 600 HRs, but rather the players themselves, Manny and Griffey (those are just his examples).

turners56
06-10-2008, 01:25 PM
The Angels only have a 40% chance of winning the AL West, while the Athletics have a 54% chance. The Angels are currently 4.5 games up.

It's funny, because in every other division the current leader is given an 80% chance to win.

If you look at the Pythagorean W-L record (I think they subtract unearned runs and then calculate your record), the Angels are nothing more than a couple of games above .500, while Oakland is 37-26. When they calculate this kind of stuff X W-L is often brought into it. Supposedly, the X W-L stat measures luck, whatever. Under X W-L, the Sox are supposed to be two games better than they are. And according to the Pythagorean W-L record, the White Sox are supposed to have the best record in the AL at 39-24.

kittle42
06-10-2008, 01:54 PM
I never get the pure hatred aimed at BP.

Because sports are not played on paper. Thus, no one should ever make hypos or predictions about anything ever.

DSpivack
06-10-2008, 01:57 PM
If you look at the Pythagorean W-L record (I think they subtract unearned runs and then calculate your record), the Angels are nothing more than a couple of games above .500, while Oakland is 37-26. When they calculate this kind of stuff X W-L is often brought into it. Supposedly, the X W-L stat measures luck, whatever. Under X W-L, the Sox are supposed to be two games better than they are. And according to the Pythagorean W-L record, the White Sox are supposed to have the best record in the AL at 39-24.

I've always thought that stat extremely stupid. It rewards teams for scoring tons of runs and seems to think it doesn't take skill to win a close game.

turners56
06-10-2008, 02:00 PM
I've always thought that stat extremely stupid. It rewards teams for scoring tons of runs and seems to think it doesn't take skill to win a close game.

But Bill James is the man!

kittle42
06-10-2008, 04:06 PM
But Bill James is the man!

As is Pythagorus!

spiffie
06-10-2008, 04:10 PM
As is Pythagorus!
Does he know Pythagoras, the famous mathematician?

kittle42
06-10-2008, 04:20 PM
Does he know Pythagoras, the famous mathematician?

Perhaps!

StepsInSC
06-10-2008, 04:53 PM
And if that happens you can be absolutely sure that you WON'T be reading a story by the stat-geeks at BP on how they got it wrong (again...)

Lip

Actually, yes, you will. It just won't be written the way you want. BP is continually looking back at their projections and what actually happened. They just don't start off by saying "WE ****ED UP" because it's entirely superfluous.

spiffie
06-10-2008, 05:08 PM
Actually, yes, you will. It just won't be written the way you want. BP is continually looking back at their projections and what actually happened. They just don't start off by saying "WE ****ED UP" because it's entirely superfluous.
The only way they could write it that would be the way the folks here want it would be "We suck, please drown us in the spat out tobacco juice of wise old scouts, we shall never write again. In fact, we are cutting off our typing fingers just to be safe."

voodoochile
06-10-2008, 05:15 PM
And if that happens you can be absolutely sure that you WON'T be reading a story by the stat-geeks at BP on how they got it wrong (again...)

Lip

Hell the folks who think they walk on water will skip 2008 altogether and crow about how they nailed 2007 on the button. It'll become the equivalent of 1969 to flubbie fans in the long run...

Lip Man 1
06-10-2008, 05:29 PM
Voodoo:

Hell I'm still waiting for their "excuses" over how the Twins could win three divisional crowns (although their Pythagorean Theory said they couldn't), how the Sox won the World Series in 2005 and how Arizona could get to the NLCS last year despite being outscored for the season! (Their chief imbecile Joe Sheehan wrote in S.I. they couldn't possibly get that far if that "trend" continued...)

:D:

For supposedly "smart" folks they sure are pretty stupid aren't they?

:D:

Lip

Optipessimism
06-10-2008, 05:42 PM
I've always thought that stat extremely stupid. It rewards teams for scoring tons of runs and seems to think it doesn't take skill to win a close game.
Exactly.

The Angels are a team built to manufacture runs and do just enough to win with their pitching.

I understand the stuff like batting average, OBP, SLG%, SB, ERA, WHIP, BB/9, K/9, BAA, BA/RISP, and all those basic stats are important, but seriously, how far beyond those basic stats do you need to go? If you want to know something about a hitter, watch some video and check out his hit chart. Just put the ****ing calculators down and go home.

Stat geeks will never understand baseball because there is no way to predict it and no way to appreciate it on paper.

On any certain pitch, you have to factor in what the managers are thinking, what the scouting reports say, the previous sequence to that hitter, the baserunners and their speed, your catchers arm, who's calling the game, the confidence the pitcher has in a certain pitch that's been called for, the weather conditions, the stadium, the hitter at the plate and whether he's hot or cold and his comfort level against the pitcher on the mound, the time of the day and the hitters backdrop and whether shadows or sun are making it hard to pick up the ball, the guy on deck, how many outs there are, what inning, the score, who is in the pen, does the pitcher have a favorable match-up, where the fielders are positioned, and you can go on and on and on and on and on. There are so many things that go on with every single pitch in a baseball game, and any one of those things mentioned can affect the outcome of a pitch which can affect the outcome of the entire game, because baseball as they say is indeed a game of inches.

Yet, hardly any of this will ever show up in the stats. The stats just group all of these completely different game situations together and then add up the outcomes and take averages. And overall the stats are a great way to predict some round about figures of what you expect a player to do overall, but when you try to group all these different situations together to make a a prediction of a player, and then you take all of these player predictions and attempt to come up with an overall number of wins for a team, it's just moronic. It's pointless. As has been mentioned and seemingly is always mentioned in these threads, literally throwing **** at the wall would prove to be just as accurate. If you gather up about twenty different baseball people and have them predict the record of a team, and you write all those predictions on a wall, and then you take a dump in your hand, close your eyes, spin around really fast and let it fly, the prediction that **** hits is just as probable as whatever figure is put out there by a bunch of nerds pounding away on their calculators.

Eddo144
06-10-2008, 06:02 PM
Stat geeks will never understand baseball because there is no way to predict it and no way to appreciate it on paper.

On any certain pitch, you have to factor in what the managers are thinking, what the scouting reports say, the previous sequence to that hitter, the baserunners and their speed, your catchers arm, who's calling the game, the confidence the pitcher has in a certain pitch that's been called for, the weather conditions, the stadium, the hitter at the plate and whether he's hot or cold and his comfort level against the pitcher on the mound, the time of the day and the hitters backdrop and whether shadows or sun are making it hard to pick up the ball, the guy on deck, how many outs there are, what inning, the score, who is in the pen, does the pitcher have a favorable match-up, where the fielders are positioned, and you can go on and on and on and on and on. There are so many things that go on with every single pitch in a baseball game, and any one of those things mentioned can affect the outcome of a pitch which can affect the outcome of the entire game, because baseball as they say is indeed a game of inches.
Of course you need to watch things to actually get a better understanding. But numbers also help to give a better understanding.

Have you ever actually read anything from BP? Or better yet, The Hardball Times? THT (www.hardballtimes.com) writes plenty of articles with statistical research. You know what else they include in some relevant articles? Videos of pitchers and their release points, to compare to previous years or other pitchers. No one from BP or THT (at least that I'm aware of), has ever said you don't need to watch players. It's just that, in order to judge the game on a more national scale, you need numbers. It's logistically impossible to break down every single player and team simply by watching. If you did that, you'd have no time to write any articles or report anything to your GM or whatever. There's a balance.

Please, read this article to see how sabermetrics helps to properly judge a player's contribution to winning: http://www.hardballtimes.com/main/article/eye-of-newt-powdered-bat-wings-and-incantations/. The author even states he's on the fence in the sabermetrics-vs-old-school debate. Then he goes on to give a valid example of how useful sabermetrics can be.

You also have to remember, projection systems aren't going to be able to predict single game outcomes. They can, however, show trends and what a team should do well and struggle with during a season. If there were such a thing as a perfect system, Vegas would not exists, as someone would have bankrupted them by now.

jabrch
06-10-2008, 06:11 PM
Of course you need to watch things to actually get a better understanding. But numbers also help to give a better understanding.



Numbers give you a good understanding of what happened. They are extremely limited in their ability to tell you what is going to happen. People are not like machines. Getting to 6 sigma percision with statistical analysis is great for manufacturing equipment. It is really poor for human performances - and even worse for complex human teams where there are so many dynamics that go unaccounted for.

Statistics are a good way to review the history of the game - provided they are taken in the right context; they just are terrible to predict the future. And there is a major difference between a complex evaluation using both stats and scouts like any good baseball team does and the hooey we get from predictive models that don't incorporate anything else and are said to be accurate in isolation.

Eddo144
06-10-2008, 06:53 PM
Numbers give you a good understanding of what happened. They are extremely limited in their ability to tell you what is going to happen. People are not like machines. Getting to 6 sigma percision with statistical analysis is great for manufacturing equipment. It is really poor for human performances - and even worse for complex human teams where there are so many dynamics that go unaccounted for.

Statistics are a good way to review the history of the game - provided they are taken in the right context; they just are terrible to predict the future. And there is a major difference between a complex evaluation using both stats and scouts like any good baseball team does and the hooey we get from predictive models that don't incorporate anything else and are said to be accurate in isolation.
Excellent post. I agree. I didn't mean to tout projection systems in my last post, just to point out their limitations. For the most part, I don't even look at PECOTA and such before the season.

Stats are most useful in judging a player's contributions. They are vital in determining who should be MVP, in the Hall of Fame, and so forth. However, you have to remember that the fun in watching the games is not knowing what will happen. Contrary to popular belief (here at least), the writers for sites like BP actually like to watch baseball. If they didn't, they'd be using their mathematical and statistical skills in a much more lucrative industry, like finance. They founded BP for the same reason you all are posting on a White Sox message board - you can never get too much baseball. :D:

Seriously, though, I highly recommend www.hardballtimes.com; the writers do a good job of analysis using both numbers and scouting reports.

Daver
06-10-2008, 06:56 PM
I don't need a bunch of numbers to know that Ken Griffey Jr. is a first ballot hall of famer.

chaerulez
06-10-2008, 07:21 PM
For all those who argue that BP and sabermetrics are trying to make the game only about numbers or throw out the line "baseball isn't played on a computer," here's a nice article (http://sportsillustrated.cnn.com/2008/baseball/mlb/05/30/bp.500HRs/index.html?eref=T1) written by Joe Sheehan (of BP) for Sports Illustrated. Sheehan writes how you shouldn't celebrate numbers, like 500 and 600 HRs, but rather the players themselves, Manny and Griffey (those are just his examples).

I always thought our obession for nice round numbers is silly. The whole 500 HR club automatic HOF entry thing was always a little foolish.

Optipessimism
06-10-2008, 07:22 PM
Of course you need to watch things to actually get a better understanding. But numbers also help to give a better understanding.

Have you ever actually read anything from BP? Or better yet, The Hardball Times? THT (www.hardballtimes.com (http://www.hardballtimes.com)) writes plenty of articles with statistical research. You know what else they include in some relevant articles? Videos of pitchers and their release points, to compare to previous years or other pitchers. No one from BP or THT (at least that I'm aware of), has ever said you don't need to watch players. It's just that, in order to judge the game on a more national scale, you need numbers. It's logistically impossible to break down every single player and team simply by watching. If you did that, you'd have no time to write any articles or report anything to your GM or whatever. There's a balance.

Please, read this article to see how sabermetrics helps to properly judge a player's contribution to winning: http://www.hardballtimes.com/main/article/eye-of-newt-powdered-bat-wings-and-incantations/. The author even states he's on the fence in the sabermetrics-vs-old-school debate. Then he goes on to give a valid example of how useful sabermetrics can be.

You also have to remember, projection systems aren't going to be able to predict single game outcomes. They can, however, show trends and what a team should do well and struggle with during a season. If there were such a thing as a perfect system, Vegas would not exists, as someone would have bankrupted them by now.

I read that article and it all seems like drivel to me.

First, the guy says that "using a sabermetric measure, such as Runs Created, ((hits+walks)(total bases)/(AB+BB)) we can discern ... the most productive player." Well I have to disagree with that. You're just taking the same stats and using them to make a different number which you say is a better way to compare players. You're not adding anything new to the equation. You're not looking at when those hits and walks are coming and whether they turn out to be meaningful in a game. It's just another empty number.

Secondly, this guy goes on to compare Carlos Baerga in 1994 (442 AB, .314 AVG, 19 HR, 80 RBI) with Max Bishop in 1930 (441 AB, .252 AVG, 10 HR, 38 RBI). See, to me this is the height of stupidity and ignorance towards the game of baseball that make most of the statheads annoying. The guy says Bishop was the more productive player because he walked 128 times to Baerga's 10 and because when you add up his runs total and RBI total, then subtract his HR total, the number is bigger than Baerga's total. Of course era doesn't count for anything, just these numbers.

If you want to dig sooooo deeply into stats that you are comparing irrelevant **** just for the fun of it, then what you're talking about has no practical use. But if you want to do it anyway, you talk about all those things that never made it on the stats. Like of course the era, the ball park size, mound height, pitcher quality, racism affecting talent level at the big leagues, etc. and then you look at the records of the teams, league averages, and most importantly clutch performances and you ask yourself "if Baerga played in that era, what would he have done," and vice versa. And you ask yourself also where each player was hitting, what each player's role on the team was, and how did he compare overall to other players in those roles. How did he contribute to his team? The only thing that you don't do is decide to say "gee, player A walked more in 1930 than player B did in 1994, so player A was more productive." That's ridiculous.

I don't have a problem with stats as far as measuring a player's season-by-season, game-by-game accomplishments. They are enormously effective and the free agent market wouldn't work without it. The problem is that when you take these stats into account, you are again ignoring the situations they come from. You're gathering up all the effects and paying no attention to the causes. So then when you try to gather up all this data and use it to come up with some asinine formula for calculating a respectable prediction of win total, you are not only ignoring the circumstances that produce those results your data is based upon, but you are also doing it 25 times over and in the process you are completely ignoring the team aspect of the game. So when you do that, you are essentially ignoring everything that makes baseball baseball, and that's why it's stupid.

Let's consider something quickly. Say you have Hitter A facing Pitcher B, a closer. Let's say it's a hot summer day, bases are loaded, and it's a 2-1 ballgame in the bottom of the 9th with 2 outs. It's a 3-1 count, Hitter A is a homerun hitter, there's nowhere to put him so you know a fastball is coming on the fastball count. In this situation, Hitter A is in a slump. He's gone like 0 for his last 23. He gets a fastball right down the middle, and he pops it up to end the game. Now in that particular situation, because the hitter is doing sooo bad, probably 90% of the time he's going to make the last out of the game and his team is going to lose. But let's consider the same exact situation except that now that hitter is hot. Probably 60% of the time, in that scenario, he's going to get a hit and win the game.

The reason I drew up that situation is because let's say both situations occur. On one occasion the hitter pops up and his team loses, and on the next he gets a hit and they win. Statistically he'd be a .500 hitter in game-on-the-line situations against that closer, but nowhere in that stat do you have enough information to predict what would likely happen if he were to come up in that situation a third time a month later. It's not a 50/50 chance he wins the game. It's a 90% chance he doesn't if he's not feeling good at the plate, or a 60% chance he does if he is feeling good at the plate.

That right there is exactly why you don't predict **** in baseball. One pitch, one swing of the bat, whatever it is, can change the entire complexion of a ballgame. Anything can happen, and in baseball it usually does. Predictions are pointless and a waste of time when they consume the type of effort BP and co. puts into them. The Red Sox didn't win last year and in 2004 because they had Bill James. They won because they had great pitching, a fearsome middle of the order, enough balance, and most importantly they got hot at the right time. By contrast, the Oakland A's haven't won a WS in a while because they didn't have enough pitching, they didn't have enough potency or balance, and they didn't get hot enough at the right times. Bill James has nothing to do with that either.

StepsInSC
06-10-2008, 09:36 PM
Voodoo:

Hell I'm still waiting for their "excuses" over how the Twins could win three divisional crowns (although their Pythagorean Theory said they couldn't), how the Sox won the World Series in 2005 and how Arizona could get to the NLCS last year despite being outscored for the season! (Their chief imbecile Joe Sheehan wrote in S.I. they couldn't possibly get that far if that "trend" continued...)

:D:

For supposedly "smart" folks they sure are pretty stupid aren't they?

:D:

Lip

It's even more stupid how you choose to ignore the obvious premise inherent in the use of past statistics to predict future events, which is that it is an uncertain endeavor at best. There are no guarantees and no certainties. Instead there are simply predictions. Most people who read Sheehan's statement about the D-Backs last year probably grasped that.

You seem to want them to eat crow every time one of their predictions is wrong. First off, they do. They point out how their predictions were wrong and analyze why. Yes I will admit they do it in an accurate way. I guess it just never really bothered me. Second, why should they have to come out and scream "we were wrong"? Everyone and their mother can understand that statistical analysis can never be used to predict the future with 100% certainty. And I know you and Daver can and do understand it, but why choose to ignore it in this situation? You're the ones effectively saying "if you can't predict with 100% certainty, then don't even bother."

BP readily and openly acknowledges that there are many other factors out there aside from past performance. Their typical shorthand for these factors is the word "luck". They use the word to refer to things that numbers can't predict. And in my opinion they openly embrace the fact that numbers cannot predict everything. I've read many times where BP has talked to scouts to discuss, for example, a pitcher's mechanics in order to ascertain why a certain pitcher may be struggling.

And how do you expect to get their "excuses" for those occurrences when you don't read BP? Are you waiting on a personal letter from the BP staff? Within 2 minutes of searching the site I found an article on the 2007 Dbacks by Sheehan (http://www.baseballprospectus.com/article.php?articleid=6597). He simply makes observations based on the numbers. He points out the Dbacks W-L record in one-run games, and how it's indicative of how great their bullpen was and how good of a job the manager did at using the right guy in the right situation. He also uses numbers to highlight how rare it is for a team to have a winning record when they are outscored and how it is "unlikely" to continue.

Why does this offend you so? It's just another means of looking at and appreciating the game of baseball. Are you secretly Buzz Bissinger? Would you be satisfied if they put up a disclaimer at the top of every article or post that said in big bold print: "THE FOLLOWING CONSISTS OF OBSERVATIONS, ANALYSIS AND EDIFICATIONS BASED ON PAST STATISTICS AND CANNOT BE USED TO PREDICT THE FUTURE TO A ASCERTAINABLE LEVEL OF CERTAINTY. THE FOLLOWING IS NOT A GURANTEE. WE HEREBY FULLY DISCLAIM ANY EXPLICIT OR IMPLICIT GUARANTEE OF CERTAINTY. PLEASE DO NOT WAGER YOUR LIFE SAVINGS ON THE STATEMENTS FOUND HEREIN." It seems to me that's what you want.

Eddo144
06-10-2008, 10:23 PM
I don't need a bunch of numbers to know that Ken Griffey Jr. is a first ballot hall of famer.
Did you even read that article? It's basically stating the opposite, that we our memories of Griffey should not be about numbers.

StepsInSC
06-10-2008, 10:25 PM
I read that article and it all seems like drivel to me.

First, the guy says that "using a sabermetric measure, such as Runs Created, ((hits+walks)(total bases)/(AB+BB)) we can discern ... the most productive player." Well I have to disagree with that. You're just taking the same stats and using them to make a different number which you say is a better way to compare players. You're not adding anything new to the equation. You're not looking at when those hits and walks are coming and whether they turn out to be meaningful in a game. It's just another empty number.

Most stats incorporate other stats. So batting average "adds nothing" more than the simple stat of at-bats?


Secondly, this guy goes on to compare Carlos Baerga in 1994 (442 AB, .314 AVG, 19 HR, 80 RBI) with Max Bishop in 1930 (441 AB, .252 AVG, 10 HR, 38 RBI). See, to me this is the height of stupidity and ignorance towards the game of baseball that make most of the statheads annoying. The guy says Bishop was the more productive player because he walked 128 times to Baerga's 10 and because when you add up his runs total and RBI total, then subtract his HR total, the number is bigger than Baerga's total. Of course era doesn't count for anything, just these numbers. What exactly is the "height of stupidity?" The cross comparing of eras? But...wouldn't the argument you're alluding to suggest that Baerga's numbers were more inflated than Bishop's because Baerga played in an more offense-inflated era? And that, therefore, the marginal value of Bishop's stats is higher than that of Baerga's? Doesn't this support the author's stance?

If you want to dig sooooo deeply into stats that you are comparing irrelevant **** just for the fun of it, then what you're talking about has no practical use. But if you want to do it anyway, you talk about all those things that never made it on the stats. Like of course the era, the ball park size, mound height, pitcher quality, racism affecting talent level at the big leagues, etc. and then you look at the records of the teams, league averages, and most importantly clutch performances and you ask yourself "if Baerga played in that era, what would he have done," and vice versa. And you ask yourself also where each player was hitting, what each player's role on the team was, and how did he compare overall to other players in those roles. How did he contribute to his team? The only thing that you don't do is decide to say "gee, player A walked more in 1930 than player B did in 1994, so player A was more productive." That's ridiculous.He didn't really say that. He pointed out Bishop's OBP was higher, and walks are a part of that. And the things you mentioned about the different aspects of eras are nothing more than platitudes that I'm sure the author, like everyone else, already accepts. In fact, the author acknowledged as such with the very last line of the article: "We may never fully understand "clutch", "intangibles", and "chemistry," but a team that acknowledges their importance will have an edge over a club with roughly equal talent that doesn’t." Isn't this paying homage to the unquantifiable qualities you just mentioned?


I don't have a problem with stats as far as measuring a player's season-by-season, game-by-game accomplishments. They are enormously effective and the free agent market wouldn't work without it. The problem is that when you take these stats into account, you are again ignoring the situations they come from. You're gathering up all the effects and paying no attention to the causes. So then when you try to gather up all this data and use it to come up with some asinine formula for calculating a respectable prediction of win total, you are not only ignoring the circumstances that produce those results your data is based upon, but you are also doing it 25 times over and in the process you are completely ignoring the team aspect of the game. So when you do that, you are essentially ignoring everything that makes baseball baseball, and that's why it's stupid.I take it you subscribe to the "if you can't predict 100% accurately, don't bother" school of reasoning. Personally, I think that's stupid. Stats should not be relied on exclusively, and I don't think anyone here has argued that they should be.


Let's consider something quickly. Say you have Hitter A facing Pitcher B, a closer. Let's say it's a hot summer day, bases are loaded, and it's a 2-1 ballgame in the bottom of the 9th with 2 outs. It's a 3-1 count, Hitter A is a homerun hitter, there's nowhere to put him so you know a fastball is coming on the fastball count. In this situation, Hitter A is in a slump. He's gone like 0 for his last 23. He gets a fastball right down the middle, and he pops it up to end the game. Now in that particular situation, because the hitter is doing sooo bad, probably 90% of the time he's going to make the last out of the game and his team is going to lose. But let's consider the same exact situation except that now that hitter is hot. Probably 60% of the time, in that scenario, he's going to get a hit and win the game.

The reason I drew up that situation is because let's say both situations occur. On one occasion the hitter pops up and his team loses, and on the next he gets a hit and they win. Statistically he'd be a .500 hitter in game-on-the-line situations against that closer, but nowhere in that stat do you have enough information to predict what would likely happen if he were to come up in that situation a third time a month later. It's not a 50/50 chance he wins the game. It's a 90% chance he doesn't if he's not feeling good at the plate, or a 60% chance he does if he is feeling good at the plate.

That right there is exactly why you don't predict **** in baseball. One pitch, one swing of the bat, whatever it is, can change the entire complexion of a ballgame. Anything can happen, and in baseball it usually does. Predictions are pointless and a waste of time when they consume the type of effort BP and co. puts into them. The Red Sox didn't win last year and in 2004 because they had Bill James. They won because they had great pitching, a fearsome middle of the order, enough balance, and most importantly they got hot at the right time. By contrast, the Oakland A's haven't won a WS in a while because they didn't have enough pitching, they didn't have enough potency or balance, and they didn't get hot enough at the right times. Bill James has nothing to do with that either.What you seem to be saying is that stats don't account for hot/cold streaks or what may happen in a given at bat. But I don't know of anyone outside of your imagination who is predicting what will happen in a given at-bat. Rather, I know of people who predict what large sample sizes will look like. Were you (plural "you") this hostile to the bell curve in your grade school math classes? It's the same thing: predicting what the distribution of a large sample size will look like. But everyone knows it will be impossible to accurately predict what one individual value in that sample size will be.

To say that the study of statistics is a waste of time is simply dismissive and disregardful. Like so many others, the "statheads" that you are railing against who claim to be able to predict with certainty the results of an at-bat are figments of your imagination. And this argument of "if it's not 100% accurate don't bother" that so many people adhere to is simply baffling.

FarWestChicago
06-10-2008, 10:31 PM
Here we go again. :bitchslap:

Daver
06-10-2008, 10:33 PM
Did you even read that article? It's basically stating the opposite, that we our memories of Griffey should not be about numbers.

Yeah I read it. I actually subscribe to BP. That said, most of the content on that site is proppelerhead garbage that has no basis in reality.

Eddo144
06-10-2008, 10:33 PM
I read that article and it all seems like drivel to me.

First, the guy says that "using a sabermetric measure, such as Runs Created, ((hits+walks)(total bases)/(AB+BB)) we can discern ... the most productive player." Well I have to disagree with that. You're just taking the same stats and using them to make a different number which you say is a better way to compare players. You're not adding anything new to the equation. You're not looking at when those hits and walks are coming and whether they turn out to be meaningful in a game. It's just another empty number.
But what he was basically saying was that Reed Johnson scored all three times because of what Hinske did at the plate, yet Wells gets all the RBIs. If you were to only use simple stats like RBI and runs, Hinske doesn't look like he contributed much at all.

Secondly, this guy goes on to compare Carlos Baerga in 1994 (442 AB, .314 AVG, 19 HR, 80 RBI) with Max Bishop in 1930 (441 AB, .252 AVG, 10 HR, 38 RBI). See, to me this is the height of stupidity and ignorance towards the game of baseball that make most of the statheads annoying. The guy says Bishop was the more productive player because he walked 128 times to Baerga's 10 and because when you add up his runs total and RBI total, then subtract his HR total, the number is bigger than Baerga's total. Of course era doesn't count for anything, just these numbers.
Yeah, that part was kind of unnecessary. I don't know why he cherry picked those two players. I'll give you this point.

If you want to dig sooooo deeply into stats that you are comparing irrelevant **** just for the fun of it, then what you're talking about has no practical use. But if you want to do it anyway, you talk about all those things that never made it on the stats. Like of course the era, the ball park size, mound height, pitcher quality, racism affecting talent level at the big leagues, etc. and then you look at the records of the teams, league averages, and most importantly clutch performances and you ask yourself "if Baerga played in that era, what would he have done," and vice versa. And you ask yourself also where each player was hitting, what each player's role on the team was, and how did he compare overall to other players in those roles. How did he contribute to his team? The only thing that you don't do is decide to say "gee, player A walked more in 1930 than player B did in 1994, so player A was more productive." That's ridiculous.
But organizations like BP have done just that. They've compared, say, the doubles rate across decades to determine if hitting 50 doubles in 2007 is better or worse than hitting 50 doubles in 1935. True, it doesn't really help with much except comparing players across eras, but it's out there.

I don't have a problem with stats as far as measuring a player's season-by-season, game-by-game accomplishments. They are enormously effective and the free agent market wouldn't work without it. The problem is that when you take these stats into account, you are again ignoring the situations they come from. You're gathering up all the effects and paying no attention to the causes. So then when you try to gather up all this data and use it to come up with some asinine formula for calculating a respectable prediction of win total, you are not only ignoring the circumstances that produce those results your data is based upon, but you are also doing it 25 times over and in the process you are completely ignoring the team aspect of the game. So when you do that, you are essentially ignoring everything that makes baseball baseball, and that's why it's stupid.

Let's consider something quickly. Say you have Hitter A facing Pitcher B, a closer. Let's say it's a hot summer day, bases are loaded, and it's a 2-1 ballgame in the bottom of the 9th with 2 outs. It's a 3-1 count, Hitter A is a homerun hitter, there's nowhere to put him so you know a fastball is coming on the fastball count. In this situation, Hitter A is in a slump. He's gone like 0 for his last 23. He gets a fastball right down the middle, and he pops it up to end the game. Now in that particular situation, because the hitter is doing sooo bad, probably 90% of the time he's going to make the last out of the game and his team is going to lose. But let's consider the same exact situation except that now that hitter is hot. Probably 60% of the time, in that scenario, he's going to get a hit and win the game.

The reason I drew up that situation is because let's say both situations occur. On one occasion the hitter pops up and his team loses, and on the next he gets a hit and they win. Statistically he'd be a .500 hitter in game-on-the-line situations against that closer, but nowhere in that stat do you have enough information to predict what would likely happen if he were to come up in that situation a third time a month later. It's not a 50/50 chance he wins the game. It's a 90% chance he doesn't if he's not feeling good at the plate, or a 60% chance he does if he is feeling good at the plate.
Well, that's ridiculously small sample size, and no statistician would ever draw any conclusions from that. Though I personally would never use statistics, no matter the sample size, to predict how a hitter would perform in a single at-bat. In fact, I don't think BP really does this either; they try to predict full-season results (though, as I'll state later, I think predictions are overrated).

That right there is exactly why you don't predict **** in baseball. One pitch, one swing of the bat, whatever it is, can change the entire complexion of a ballgame. Anything can happen, and in baseball it usually does. Predictions are pointless and a waste of time when they consume the type of effort BP and co. puts into them. The Red Sox didn't win last year and in 2004 because they had Bill James. They won because they had great pitching, a fearsome middle of the order, enough balance, and most importantly they got hot at the right time. By contrast, the Oakland A's haven't won a WS in a while because they didn't have enough pitching, they didn't have enough potency or balance, and they didn't get hot enough at the right times. Bill James has nothing to do with that either.
I sort of agree with you here. Preseason predictions should really just be for fun. I'm more of a "stats in retrospect" type of person. And you're right that Bill James made exactly zero contribution in October for the Red Sox. However, someone like Bill James can help your organization to find the proper players to spend your resources on in the offseason.

I don't mean to start **** here, just giving my opinion. A lot of what you've cited and argued against are things that sabermetric types don't really claim to be true. I guess that's my biggest problem - people on both sides tend to get in heated arguments without actually understanding what the other side is arguing.

Eddo144
06-10-2008, 10:34 PM
Yeah I read it. I actually subscribe to BP. That said, most of the content on that site is proppelerhead garbage that has no basis in reality.
Sorry Daver, I think we're referring to different things. I was referring to this SI article (http://sportsillustrated.cnn.com/2008/baseball/mlb/05/30/bp.500HRs/index.html?eref=T1). Apologies for the blunt reply.

Lip Man 1
06-10-2008, 11:19 PM
Steps writes almost an entire full page reply, using many words with multiple syllables, that you need a doctorate in Newtonian physics to understand, to try to convince everyone how BP and the other "propellerheads" (love that term Daver), really don't mean to come across as arrogant, know it all's.

The fact that it takes almost an entire page, full of symbols, formulas and constructs to try to make his point says volumes don't you think?

It's a human game full of variables and uncertainties that no scientist, no mathematician, not even the great Albert Einstein or Dr. Stephen Hawking could try to predict.

Ladies and gentleman of the jury, the defense rests.

Lip

StepsInSC
06-10-2008, 11:25 PM
Steps writes...to convince everyone how BP and the other "propellerheads" (love that term Daver), really don't mean to come across as arrogant, know it all's.

Actually, that was explicitly not my point. And arrogance can reveal itself through over-simplification as well as over-"complification."

Eddo144
06-10-2008, 11:28 PM
Steps writes almost an entire full page reply, using many words with multiple syllables, that you need a doctorate in Newtonian physics to understand, to try to convince everyone how BP and the other "propellerheads" (love that term Daver), really don't mean to come across as arrogant, know it all's.

The fact that it takes almost an entire page, full of symbols, formulas and constructs to try to make his point says volumes don't you think?

It's a human game full of variables and uncertainties that no scientist, no mathematician, not even the great Albert Einstein or Dr. Stephen Hawking could try to predict.

Ladies and gentleman of the jury, the defense rests.

Lip
Optipessimism also wrote a tremendously long post arguing the opposite of Steps, so I don't really see your point.

Secondly, Steps and I have both said that there's no way predictions will ever be 100% accurate, so I don't see why you're still arguing that point.

Thirdly, the term "propellerhead" evokes images of beanies on nerds, I assume. Do you really still need to make fun of nerds, just because they try to find ways to better understand baseball with mathematics? Are we still in junior high?

Apparently, there's no way we'll ever get you guys to budge one inch. We're not asking you to take BP on blind faith (I sure as hell don't), just asking that you acknowledge that statistics aren't pulled out of thin air and actually have some basis in reality. Remember, there have been thousands upon thousands of innings in baseball, and situations like "1st and 3rd, 1 out, bottom of the 5th, home team up by 2" have happened enough times to where we can say that the home team will win X% of the time. BP doesn't make up that X, they get it from actual game outcomes.

fquaye149
06-10-2008, 11:32 PM
It is all about computer generated simulations. It means.... Yes.... NOTHING

wow what an insightful post!

It means nothing just like Andruw Jones's 2007 stats meant nothing in suggesting that Andruw Jones could have a bad 2008

FedEx227
06-10-2008, 11:38 PM
wow what an insightful post!

It means nothing just like Andruw Jones's 2007 stats meant nothing in suggesting that Andruw Jones could have a bad 2008

Nope. Eyes never lie.

I never understood the hate for prediction stats... they are PREDICTIONS. There is never a disclaimer that says "By the way, this is exactly how things are going to happen next year and if anything changes it's just because they beat our system or played above our system..." they never say that.

THEY ARE PREDICTIONS. Just like you and me saying Andruw Jones is going to have an awful 2008. It's really not that ground-breaking.

fquaye149
06-10-2008, 11:45 PM
Nope. Eyes never lie.

I never understood the hate for prediction stats... they are PREDICTIONS. There is never a disclaimer that says "By the way, this is exactly how things are going to happen next year and if anything changes it's just because they beat our system or played above our system..." they never say that.

THEY ARE PREDICTIONS. Just like you and me saying Andruw Jones is going to have an awful 2008. It's really not that ground-breaking.

I don't get why people can't understand that people like SABRmetrics/stats/etc because it is another way to enjoy baseball.

The Bill James historical abstract is one of my favorite books ever, not because it gives me a "forecast" for what will happen (how could it? Most of the players mentioned in the book have long retired) nor because I think it proves anything definitively, but because James loves the game, provides unbelievably intriguing anecdotes about ballplayers, the changing rules and complexion of the game, and gives you a new way of looking at traditional paradigms of what makes a player great, and reaffirms other traditional "positive aspects" of a player (for instance, Bill James is a big proponent of not GIDP, getting HBP, and being an excellent baserunner)

Daver
06-10-2008, 11:49 PM
Optipessimism also wrote a tremendously long post arguing the opposite of Steps, so I don't really see your point.

Secondly, Steps and I have both said that there's no way predictions will ever be 100% accurate, so I don't see why you're still arguing that point.

Thirdly, the term "propellerhead" evokes images of beanies on nerds, I assume. Do you really still need to make fun of nerds, just because they try to find ways to better understand baseball with mathematics? Are we still in junior high?

Apparently, there's no way we'll ever get you guys to budge one inch. We're not asking you to take BP on blind faith (I sure as hell don't), just asking that you acknowledge that statistics aren't pulled out of thin air and actually have some basis in reality. Remember, there have been thousands upon thousands of innings in baseball, and situations like "1st and 3rd, 1 out, bottom of the 5th, home team up by 2" have happened enough times to where we can say that the home team will win X% of the time. BP doesn't make up that X, they get it from actual game outcomes.

VORP is based on pure fiction.

PECOTA is based on pure fiction

The save and the hold are worthless stats that were added to the game for the simple reason of giving agents a bargaining point. Most stats that are not batting average, RBI's, HR's , and ERA are stats created to aid agents in negotiations for FA players.

Let's not get ahead of ourselves on how worthless a lot of these numbers are.

fquaye149
06-10-2008, 11:55 PM
VORP is based on pure fiction.

You mean finding out how well a player hits and fields compared to a mediocre player at his position is based on pure fiction?

:scratch:


VORP is based on Batting Average, On Base Average, and what percentage of a player's hits went for extra bases. Those figures, upon which VORP is based, sure sound closer to "fact" than "fiction"



PECOTA is based on pure fictionPECOTA's results may be pure fiction (probably not) but it is BASED ON computing things that most certainly HAVE happen


The save and the hold are worthless stats that were added to the game for the simple reason of giving agents a bargaining point. Most stats that are not batting average, RBI's, HR's , and ERA are stats created to aid agents in negotiations for FA players.That doesn't make them fiction.

The save actually exists. It may or may not be worth much as a stat but it certainly exists. It is clearly defined and people either do or do not record a save.


Let's not get ahead of ourselves on how worthless a lot of these numbers are.RBI was not always a statistic. At one point someone had to say "this is a stat that will be recorded."

Ditto for SB.

Did baseball record keeping attain perfection in 1905? I doubt it.

StepsInSC
06-11-2008, 12:06 AM
VORP is based on pure fiction.

PECOTA is based on pure fiction

The save and the hold are worthless stats that were added to the game for the simple reason of giving agents a bargaining point. Most stats that are not batting average, RBI's, HR's , and ERA are stats created to aid agents in negotiations for FA players.

Let's not get ahead of ourselves on how worthless a lot of these numbers are.

Your logic is compelling.

Eddo144
06-11-2008, 12:10 AM
VORP is based on pure fiction.

PECOTA is based on pure fiction

The save and the hold are worthless stats that were added to the game for the simple reason of giving agents a bargaining point. Most stats that are not batting average, RBI's, HR's , and ERA are stats created to aid agents in negotiations for FA players.

Let's not get ahead of ourselves on how worthless a lot of these numbers are.
VORP, just like ERA and batting average, is calculated based on what happened on the field. It is not "fiction", but rather a way of arranging the records of what has occurred.

PECOTA is BP's projection system, which tries to predict the future based on previous results. It tries to predict the future. I suppose that is sort of like fiction.

The save was created by legendary sportswriter Jerome Holtzman (sp?). No sabermetrically inclined person really respects the save - it's about as arbitrary an achievement there is. (On a side note, explain the save to someone who doesn't follow baseball*, you'll sound ridiculous. This is a good indicator that saves are fairly pointless.) Holds are basically in the same boat as saves and pitcher wins and losses. They are all arbitrarily defined.

I don't see how things like VORP were developed to "aid agents". As far as I know, BP developed VORP, and they don't represent any players in negotiations.

*"When a pitcher comes into a game when his team is up by three runs or less, and finishes said game without ever giving up the lead, he gets a save." Why is it three runs? Why not one? Why not five? Completely arbitrary.

At least VORP (and I tend to like more "concrete" metrics myself, ones that will stay the same year-by-year (the definition of "replacement player" can change over time, so a VORP of 100 in 2008 might not equal 100 ten years from now)) is easily definable: "The value, in runs, a player produces relative to what an average AAA player at his position would produce." I don't claim that it's perfectly calculated, but at least it's an admirable goal.

Daver
06-11-2008, 12:14 AM
Your logic is compelling.

It saves a lot of typing too.

I can watch someone play baseball and have a good idea whether they are good or bad at it.

Strange concept to the proppelerheads, but it has worked for over a hundred years.

Eddo144
06-11-2008, 12:15 AM
I can watch someone play baseball and have a good idea whether they are good or bad at it.

Strange concept to the proppelerheads, but it has worked for over a hundred years.
Wow. So can I.

But can you tell me how good someone you haven't seen is? It's almost as though you'd need some sort of record of what they've produced. Hmm...

fquaye149
06-11-2008, 12:28 AM
*"When a pitcher comes into a game when his team is up by three runs or less, and finishes said game without ever giving up the lead, he gets a save." Why is it three runs? Why not one? Why not five? Completely arbitrary.


I don't think you understand how a save is awarded

a save goes to a pitcher who

1. finishes a game
2. is not the winning pitcher
3. records at least one out

and one of these three conditions:

4. pitches at least one inning with a 3 run lead (3 outs in an inning, 3 runs, e.g. not arbitrary)

AND/OR

4a. comes into the game with the tying run on base, at the plate, or on deck

AND/OR

4b. pitches at least three innings

only 4b. is arbitrary, but there aren't many 3 inning+ saves.

4 and 4a make perfect sense. If the reliever comes into the game and can give up as many runs as records outs, he shouldn't get the save

Daver
06-11-2008, 12:33 AM
Wow. So can I.

But can you tell me how good someone you haven't seen is? It's almost as though you'd need some sort of record of what they've produced. Hmm...

Why speculate on something you have not seen, knowing that their is no basis in fact to your speculation?

kittle42
06-11-2008, 12:43 AM
Why speculate on something you have not seen, knowing that their is no basis in fact to your speculation?

OK, being in the middle of these two camps (I can enjoy the BP-type stuff and love old baseball stats, but also don't care about things like Pythagorean WL and, say, the 4th page of ESPN's "see further statistics" on each player)...

No one here is arguing that stats are perfect or anywhere near perfect or even above average in predicting things. But they are another fun way to enjoy the game.

Daver and Lip shout down absolutely anything having to do with the stats and refuse to give an inch at all.

It is usually the latter type of argument that is the flawed one.

Optipessimism
06-11-2008, 09:07 AM
Most stats incorporate other stats. So batting average "adds nothing" more than the simple stat of at-bats?

Batting average is a basic stat. It doesn't add anything beyond it's intended value, which is to approximate how often a hitter gets a hit. It's a simple stat that is meant to exist on its own. Once you take it out of the context it was designed for and start combing it with a bunch of other stats it loses the value it has.

What exactly is the "height of stupidity?" The cross comparing of eras? But...wouldn't the argument you're alluding to suggest that Baerga's numbers were more inflated than Bishop's because Baerga played in an more offense-inflated era? And that, therefore, the marginal value of Bishop's stats is higher than that of Baerga's? Doesn't this support the author's stance?

You're missing the point. It's easy to say that because Baerga played in the steroids era and small ballpark era his numbers were inflated so Bishop would have been more productive. But, it's also easy to say that Bishop walked more because the pitching wasn't as good, and that the walks he took are overvalued because he didn't play in an era of home runs. You additionally must consider that the lack of non-white players in the league made it much weaker, and that baseball salaries in those days weren't attractive to talented athletes with more lucrative career opportunities, and that you didn't have 5+ levels of minor leagues with generally well-trained career coaches, scouts, and player development personnel making up staffs to improve the quality of competition. Knowing this, it is very easy to suggest that had Bishop had played in Baerga's era, he never would have risen above Double A.

The arguments over eras can never be settled because the situations couldn't be more different. Compaing a player in 1930 to a player in 1994 is just like comparing a football player today to a badmitten player today. It doesn't work, and stats are never going to help.

He didn't really say that. He pointed out Bishop's OBP was higher, and walks are a part of that. And the things you mentioned about the different aspects of eras are nothing more than platitudes that I'm sure the author, like everyone else, already accepts. In fact, the author acknowledged as such with the very last line of the article: "We may never fully understand "clutch", "intangibles", and "chemistry," but a team that acknowledges their importance will have an edge over a club with roughly equal talent that doesn’t." Isn't this paying homage to the unquantifiable qualities you just mentioned?

Finding another way to rearrange numbers by ignoring the specific situations isn't paying homage to anything. He's admitting that there are factors beyond stats because he has to. You can't watch a baseball game without noticing that.

I think it's that you guys don't understand just how easy this stat **** is to do yourself.

I value walks, home runs, and players who make a lot of contact, so I'm going to make up my own stat right now. I'll call it the CANNOPEB stat (Creating a New Number Out of a Personal Bias).

I'm going to assume home runs are equal to four walks just because I can. So lets take [(HR * 4) + BB + HBP] - K. Go do that and make a list of of the league leaders in CANNOPEB. I can guarantee that some pretty good players are going to show up on the list. But, just because they do, does it mean the stat I've just created is useful? Nope.

I take it you subscribe to the "if you can't predict 100% accurately, don't bother" school of reasoning. Personally, I think that's stupid. Stats should not be relied on exclusively, and I don't think anyone here has argued that they should be.

I don't think that anything less than perfection regarding predictions is useless. Not at all.

The reality is that it is impossible for baseball statisticians to predict anything with enough accuracy to even come close to replacing the opinion of a good baseball man. So yes, the system as a whole is completely useless when you consider that the nerds not only misrepresent the game, they can't even do the job they're supposed to do.

What you seem to be saying is that stats don't account for hot/cold streaks or what may happen in a given at bat. But I don't know of anyone outside of your imagination who is predicting what will happen in a given at-bat. Rather, I know of people who predict what large sample sizes will look like. Were you (plural "you") this hostile to the bell curve in your grade school math classes? It's the same thing: predicting what the distribution of a large sample size will look like. But everyone knows it will be impossible to accurately predict what one individual value in that sample size will be.

You're not supposed to try to predict what will happen in a given at bat unless you have stats that apply to the current situation. The point is, stats are meant to exist in the past as a collection of results ignorant of the circumstances under which they originated. They are meant to be viewed only for what they are worth.

There's nothing wrong with making a graph comparing players through batting average, OBP, HR's, SLG%, ERA, WHIP, K/9, etc. There's nothing wrong with looking at a couple of graphs and saying, "Player A had a K/9 of 8.79 two years ago, a K/9 of 9.42 last year, and so far this year he has a K/9 of 11.0, so I expect him to come down a tad to around 9." The problem is mixing up all these stats and using them to try to predict something.

To say that the study of statistics is a waste of time is simply dismissive and disregardful. Like so many others, the "statheads" that you are railing against who claim to be able to predict with certainty the results of an at-bat are figments of your imagination. And this argument of "if it's not 100% accurate don't bother" that so many people adhere to is simply baffling.

I never said that basic stats viewed in proper context were a waste of time. However, taking a bunch of stats out of context and rearranging them to try to create a new stat that you think will be better is.

Go dig up the PECOTA predictions for the last few seasons and calculate exactly how wrong they've been. Then ask yourself, considering such a wide margin of error, would you EVER in your life trust such a prediction if it concerned itself with anything important to you? If a prediction based on stats is no more valuable than the guesstimates of an average fan, it has no special value whatsoever.

I have no qualms with the people that take in interest in creating new stats, nor do I have problems with BP making up a bunch of malarky. The problem is when people bring this **** into a legitimate baseball conversation like it actually proves something.

It's like, imagine if you're talking with a friend about the origin of man. Both you and your friend have studied this issue before from both sides so the conversation is pretty engaging. Now lets say some Scientologist enters the picture and says you and your friend are both wrong, and he knows better than either of you because he has a higher operating thetan level. He says that what actually happened was this big alien named Xenu threw some other aliens into a volcano and bombed them, then the alien souls entered like monkeys or something and thats how man was created. That's exactly what it feels like when you're having a serious baseball conversation and someone else brings up PECOTA and all that other garbage. It's trendy and that's about the nicest thing you can say about it.

Optipessimism
06-11-2008, 09:15 AM
Here we go again. :bitchslap:
What is the probability of a personal injury?

Optipessimism
06-11-2008, 09:24 AM
I sort of agree with you here. Preseason predictions should really just be for fun. I'm more of a "stats in retrospect" type of person. And you're right that Bill James made exactly zero contribution in October for the Red Sox. However, someone like Bill James can help your organization to find the proper players to spend your resources on in the offseason.

I don't mean to start **** here, just giving my opinion. A lot of what you've cited and argued against are things that sabermetric types don't really claim to be true. I guess that's my biggest problem - people on both sides tend to get in heated arguments without actually understanding what the other side is arguing.
I respect your argument but this highlighted portion I want to respond to.

It seems like Beane and James get a whole lot of credit for appreciating the walk-walk-three run homer approach to baseball, and deservedly so. It can be tremendously effective over a 162 game season even if it's not suited for playoff baseball against top pitching staffs.

What kind of annoys me though is that somehow, somewhere, the credit that should be going mainly to Billy Beane's scouting and player development staff gets transferred to publications like BP and systems like PECOTA because of a loosely shared philosophy.

Eddo144
06-11-2008, 09:26 AM
Why speculate on something you have not seen, knowing that their is no basis in fact to your speculation?
That's nonsense. If I say got Player X on base 36.3% of the time in 2007, that's a fact. If I say Player Y had 33 HR in 2007, that's also a fact. Just because I didn't see all 33 HR doesn't change that.

Now, I know you're not referring to PECOTA, because I've already said I don't care for PECOTA and its projections.

Craig Grebeck
06-11-2008, 09:29 AM
He says that what actually happened was this big alien named Xenu threw some other aliens into a volcano and bombed them, then the alien souls entered like monkeys or something and thats how man was created. That's exactly what it feels like when you're having a serious baseball conversation and someone else brings up PECOTA and all that other garbage. It's trendy and that's about the nicest thing you can say about it. That's honestly the dumbest thing I've ever read in my entire life.

Here is the average error in PECOTA team predictions 2003-2007:
2003 5.91 wins
2004 7.71 wins
2005 5.14 wins
2006 4.94 wins
2007 4.31 wins

In 2005 they began incorporating defense and have become much, much more accurate.

Eddo144
06-11-2008, 09:30 AM
I value walks, home runs, and players who make a lot of contact, so I'm going to make up my own stat right now. I'll call it the CANNOPEB stat (Creating a New Number Out of a Personal Bias).

I'm going to assume home runs are equal to four walks just because I can. So lets take [(HR * 4) + BB + HBP] - K. Go do that and make a list of of the league leaders in CANNOPEB. I can guarantee that some pretty good players are going to show up on the list. But, just because they do, does it mean the stat I've just created is useful? Nope.
But that's not they come up with the metrics. They don't say, "I'm going to assume..." A more basic stat, runs created (times on base * total bases)/(plate appearances), was accidentally stumbled upon by Bill James when he discovered it correlated very well with the actual number of runs a team scored in any given year.

Eddo144
06-11-2008, 09:32 AM
I respect your argument but this highlighted portion I want to respond to.

It seems like Beane and James get a whole lot of credit for appreciating the walk-walk-three run homer approach to baseball, and deservedly so. It can be tremendously effective over a 162 game season even if it's not suited for playoff baseball against top pitching staffs.

What kind of annoys me though is that somehow, somewhere, the credit that should be going mainly to Billy Beane's scouting and player development staff gets transferred to publications like BP and systems like PECOTA because of a loosely shared philosophy.
You have a good point, but that's not BP's fault, more of the fault of the media and readers who don't quite understand what's going on.

Personally, I'd be shocked if either Beane and the A's or James and the Red Sox actually use BP's stats. My guess is that they have their own metrics.

And both teams still use scouts and such. The whole reason Beane started to use sabermetrics was because scouts are so expensive, and the A's operate under quite a tight budget. He figured that he could use something free (numbers) to replace some of his more expensive resources (scouts).

Optipessimism
06-11-2008, 09:34 AM
But that's not they come up with the metrics. They don't say, "I'm going to assume..." A more basic stat, runs created (times on base * total bases)/(plate appearances), was accidentally stumbled upon by Bill James when he discovered it correlated very well with the actual number of runs a team scored in any given year.
Okay, well you guys win. I don't wanna debate this anymore.

Craig Grebeck
06-11-2008, 09:39 AM
What kind of annoys me though is that somehow, somewhere, the credit that should be going mainly to Billy Beane's scouting and player development staff gets transferred to publications like BP and systems like PECOTA because of a loosely shared philosophy.
You do realize that BP acknowledges the brilliance of A's scouting? Goldstein just wrote in a chat last week about their strong scouting.


It seems like Beane and James get a whole lot of credit for appreciating the walk-walk-three run homer approach to baseball, and deservedly so. It can be tremendously effective over a 162 game season even if it's not suited for playoff baseball against top pitching staffs.
Yeah -- like the Red Sox vs. Angels series last season where the Red Sox thumped them while taking 16 BB in three games. Or the Yankees vs. Indians series where the Indians took 20 BB in four games. Or when the D'backs took 13 BB in three games against the Cubs. The Rockies took the least amount, eleven, against the Phillies.

The ALCS: Red Sox took 31 BB in seven games.
The NLCS: Rockies took 18 BB in four games.

WS: Red Sox took 19 BB in four games.

I don't think the Red Sox would agree with your statement.

Optipessimism
06-11-2008, 09:44 AM
You do realize that BP acknowledges the brilliance of A's scouting? Goldstein just wrote in a chat last week about their strong scouting.


Yeah -- like the Red Sox vs. Angels series last season where the Red Sox thumped them while taking 16 BB in three games. Or the Yankees vs. Indians series where the Indians took 20 BB in four games. Or when the D'backs took 13 BB in three games against the Cubs. The Rockies took the least amount, eleven, against the Phillies.

The ALCS: Red Sox took 31 BB in seven games.
The NLCS: Rockies took 18 BB in four games.

WS: Red Sox took 19 BB in four games.

I don't think the Red Sox would agree with your statement.

Those walk numbers don't sound like the products of great pitching.

Optipessimism
06-11-2008, 09:46 AM
You have a good point, but that's not BP's fault, more of the fault of the media and readers who don't quite understand what's going on.

Personally, I'd be shocked if either Beane and the A's or James and the Red Sox actually use BP's stats. My guess is that they have their own metrics.

And both teams still use scouts and such. The whole reason Beane started to use sabermetrics was because scouts are so expensive, and the A's operate under quite a tight budget. He figured that he could use something free (numbers) to replace some of his more expensive resources (scouts).
Beane is a very smart guy and he does a great job with the budget he has. The Twins are the same way, only they value things differently. Carlos Gomez would never lead off for the A's for example.

Craig Grebeck
06-11-2008, 09:50 AM
Those walk numbers don't sound like the products of great pitching.
Well, I thought walking in the postseason was not likely due to strong pitching.

kittle42
06-11-2008, 10:57 AM
Compaing a player in 1930 to a player in 1994 is just like comparing a football player today to a badmitten player today. It doesn't work, and stats are never going to help.

I'd love to see this game played live! :D:

Lip Man 1
06-11-2008, 01:20 PM
Kittle:

Actually I like baseball statistics....they are a part of the game and provide interesting talking points. (They also provide needed filler on the back of a baseball card! - LOL!)

I like the statistics that are meaningful and easily understood, wins, losses, hits, ERA, batting average for example...that have a place in everyday baseball.

What I don't like are the people who think statistics are the be all and end all to everything, who try to reduce a magnificent game to simple mathematics that can be discussed in a classroom. That try to take the human element out of things. That "invent" meaningless statistics, that you need an advanced degree in physics to understand, and that basically are irrelevant to the realities of the game.

I also don't like the attitude of some of the stat sites that crow to the heavens when they get one thing dead on (by pure luck) yet are mysteriously silent when they get a "prediction" completely wrong and I've pointed out five examples just in the past seven years or so.

Fair is fair right? If you are going to make a big deal (aka Dan Van Dyke of the Tribune) that BP got the Sox 2007 record "right," then where's the story that they looked completely stupid in 2005?

Here's a mathematical premise that's easy to understand and dead on. When there are billions of variables in just one single baseball game that help decide who wins and who loses, it's a pretty good idea to stay away from "predicting" seasonal outcomes which as other have said rely on using numbers from the past as a means to explain what is going to happen today or tomorrow or next month.

What the stat-heads refuse to admit is that despite all their fancy names and mathematical constructs they are simply GUESSING....just like everyone one else.

Fed Ex is right, those sites aren't saying "this is the way things WILL be..." so then why don't they just use that term in their pages and pages of notes "this is an educated GUESS."

Come on BP, use the term...say exactly what it is in reality... a GUESS. But then if they did that some of their "geniuses" would lose some of their luster wouldn't it?

Lip

Daver
06-11-2008, 01:35 PM
That's honestly the dumbest thing I've ever read in my entire life.

Here is the average error in PECOTA team predictions 2003-2007:
2003 5.91 wins
2004 7.71 wins
2005 5.14 wins
2006 4.94 wins
2007 4.31 wins

In 2005 they began incorporating defense and have become much, much more accurate.

And it is still pure mental masturbation.

spiffie
06-11-2008, 01:46 PM
It saves a lot of typing too.

I can watch someone play baseball and have a good idea whether they are good or bad at it.

Strange concept to the proppelerheads, but it has worked for over a hundred years.
If it's that simple why was Joe Borchard a multimillionaire before he ever stepped foot on a major league baseball field?

Daver
06-11-2008, 02:01 PM
If it's that simple why was Joe Borchard a multimillionaire before he ever stepped foot on a major league baseball field?


Ask Kenny Williams, he had him scouted not me.

spiffie
06-11-2008, 02:04 PM
Ask Kenny Williams, he had him scouted not me.
Okay, I misunderstood your point. Telling who is a good and bad baseball player is easy for you, as opposed to MLB organizations who continually make a lot of crappy decisions. I thought your implication is that simply watching a baseball player once or twice is plenty to make that judgment. Mea Culpa.

fquaye149
06-11-2008, 02:45 PM
Kittle:

Actually I like baseball statistics....they are a part of the game and provide interesting talking points. (They also provide needed filler on the back of a baseball card! - LOL!)

I like the statistics that are meaningful and easily understood, wins, losses, hits, ERA, batting average for example...that have a place in everyday baseball.

What I don't like are the people who think statistics are the be all and end all to everything, who try to reduce a magnificent game to simple mathematics that can be discussed in a classroom. That try to take the human element out of things. That "invent" meaningless statistics, that you need an advanced degree in physics to understand, and that basically are irrelevant to the realities of the game.

I also don't like the attitude of some of the stat sites that crow to the heavens when they get one thing dead on (by pure luck) yet are mysteriously silent when they get a "prediction" completely wrong and I've pointed out five examples just in the past seven years or so.

Fair is fair right? If you are going to make a big deal (aka Dan Van Dyke of the Tribune) that BP got the Sox 2007 record "right," then where's the story that they looked completely stupid in 2005?

Here's a mathematical premise that's easy to understand and dead on. When there are billions of variables in just one single baseball game that help decide who wins and who loses, it's a pretty good idea to stay away from "predicting" seasonal outcomes which as other have said rely on using numbers from the past as a means to explain what is going to happen today or tomorrow or next month.

What the stat-heads refuse to admit is that despite all their fancy names and mathematical constructs they are simply GUESSING....just like everyone one else.

Fed Ex is right, those sites aren't saying "this is the way things WILL be..." so then why don't they just use that term in their pages and pages of notes "this is an educated GUESS."

Come on BP, use the term...say exactly what it is in reality... a GUESS. But then if they did that some of their "geniuses" would lose some of their luster wouldn't it?

Lip


Well BP aren't those people, so your angst is misdirected.

Why don't they say "This is an educated guess"? Because it's understood to be that by people who don't completely misinterpret what it's meant to be

Oblong
06-11-2008, 02:47 PM
I like the statistics that are meaningful and easily understood, wins, losses, hits, ERA, batting average for example...that have a place in everyday baseball.

What I don't like are the people who think statistics are the be all and end all to everything, who try to reduce a magnificent game to simple mathematics that can be discussed in a classroom. That try to take the human element out of things. That "invent" meaningless statistics, that you need an advanced degree in physics to understand, and that basically are irrelevant to the realities of the game.

I also don't like the attitude of some of the stat sites that crow to the heavens when they get one thing dead on (by pure luck) yet are mysteriously silent when they get a "prediction" completely wrong and I've pointed out five examples just in the past seven years or so.

Fair is fair right? If you are going to make a big deal (aka Dan Van Dyke of the Tribune) that BP got the Sox 2007 record "right," then where's the story that they looked completely stupid in 2005?

Here's a mathematical premise that's easy to understand and dead on. When there are billions of variables in just one single baseball game that help decide who wins and who loses, it's a pretty good idea to stay away from "predicting" seasonal outcomes which as other have said rely on using numbers from the past as a means to explain what is going to happen today or tomorrow or next month.

What the stat-heads refuse to admit is that despite all their fancy names and mathematical constructs they are simply GUESSING....just like everyone one else.



Lip

well I wouldn't talk about meaningfull stats and then throw out Wins, Losses, and ERA if you are referring to measuring a pitcher. Those are probably the 3 least meaningful ways to measure one. It's not 1935 anymore. It doesn't take into account team defense or team offense. It's that reasoning why Bert Blyleven's not in the HOF.

And isn't a prediction, by definition, a guess? There is no difference so I don't get your overall point.

kittle42
06-11-2008, 03:46 PM
Well BP aren't those people, so your angst is misdirected.

Why don't they say "This is an educated guess"? Because it's understood to be that by people who don't completely misinterpret what it's meant to be

Totally agree. You just saved me a longer post.

Lip Man 1
06-11-2008, 04:44 PM
Kittle:

I may be dead wrong on this regarding the individual and organization but if you can somehow look through the WSI archives from a few years ago you'll find a thread quoting a post from one of these places that was written by the "publisher" (and published on their web site) outright condemning Sox fans because they had the "gall" to e-mail, write and call by the hundreds to said organization ripping the **** out of them because of their stupidity with the Sox in 2005.

I swear it was Sheehan and BP but I could be wrong on the individual and organization and if so I'll admit I was wrong but I KNOW that thread is somewhere here at WSI.

If anyone else remembers this specific denunciation, I'd appreciate you chiming in on this. I know it got some responses.

It had to do with, if memory serves, that site, AFTER the Sox had clinched the division (which means they had a 100% chance of making the post season) they CONTINUED to say they had an 88% chance or a 94% chance of doing so.

THAT above all else turned me completely and utterly against those arrogant, egotistical morons. The fact that when they were called out over it, instead of admitting they were wrong and saying it wouldn't happen again, they (or one individual in authority) had the gonads to "blame" Sox fans.

Lip

Lip Man 1
06-11-2008, 04:50 PM
Oblong:

I like those stats because they are simple and easy to understand. I don't need a degree in quantum mechanics to figure out or even try to understand what some of these people are trying to say.

VORP and OBP and XYZ and most of that other crap is meaningless gobbeldegook unless you understand advanced calculus.

If something is that hard to fathom it's usually not relevant to average people.

Besides I could care less what Jermaine Dye hit against left handed pitchers on the 3rd Thursday of every month, on the road, in temperatures at least 70 degrees.

Many of these "stats" are that obtuse and goofy.

Lip

Eddo144
06-11-2008, 04:54 PM
It had to do with, if memory serves, that site, AFTER the Sox had clinched the division (which means they had a 100% chance of making the post season) they CONTINUED to say they had an 88% chance or a 94% chance of doing so.
OK, their response to that may not have been very professional, but to completely discount a site because their playoff odds were slightly off due to some minor mathematical error is ridiculous.

Remember, the playoff odds are objective. They plug in a bunch of input variables (current record, runs scored, runs against, etc.), and the program spits out odds. Obviously, something was wrong with the program at that time where it didn't recognize an actual clinching of the division (I believe it didn't do the tiebreakers correctly).

They didn't keep the Sox at a lower percentage because they had a vendetta. I bet a lot of the BP hatred on this site would be different if they had become popular during a time where the Sox weren't such at outlier. Oh well.

Eddo144
06-11-2008, 04:57 PM
Oblong:

I like those stats because they are simple and easy to understand. I don't need a degree in quantum mechanics to figure out or even try to understand what some of these people are trying to say.

VORP and OBP and XYZ and most of that other crap is meaningless gobbeldegook unless you understand advanced calculus.

If something is that hard to fathom it's usually not relevant to average people.

Besides I could care less what Jermaine Dye hit against left handed pitchers on the 3rd Thursday of every month, on the road, in temperatures at least 70 degrees.

Many of these "stats" are that obtuse and goofy.

Lip
Sure, VORP is complicated, mainly because it's computed relative to a replacement level, which can change over time.

But OBP, seriously? OBP is the exact same calculation as batting average, except it's times on base/plate appearances instead of hits/at bats. Saying that OBP is too complicated is really showing your ignorance.

And the kind of stats BP uses are the exact opposite of "what Jermaine Dye hit against left handed pitchers on the 3rd Thursday of every month, on the road, in temperatures at least 70 degrees." Your example is cherry picking, and reducing the relevant sample size to virtually nothing. True statistical analysis requires a huge sample size - many seasons worth of data.

fquaye149
06-11-2008, 05:02 PM
Kittle:

I may be dead wrong on this regarding the individual and organization but if you can somehow look through the WSI archives from a few years ago you'll find a thread quoting a post from one of these places that was written by the "publisher" (and published on their web site) outright condemning Sox fans because they had the "gall" to e-mail, write and call by the hundreds to said organization ripping the **** out of them because of their stupidity with the Sox in 2005.

I swear it was Sheehan and BP but I could be wrong on the individual and organization and if so I'll admit I was wrong but I KNOW that thread is somewhere here at WSI.

If anyone else remembers this specific denunciation, I'd appreciate you chiming in on this. I know it got some responses.

It had to do with, if memory serves, that site, AFTER the Sox had clinched the division (which means they had a 100% chance of making the post season) they CONTINUED to say they had an 88% chance or a 94% chance of doing so.

THAT above all else turned me completely and utterly against those arrogant, egotistical morons. The fact that when they were called out over it, instead of admitting they were wrong and saying it wouldn't happen again, they (or one individual in authority) had the gonads to "blame" Sox fans.

Lip

They had clinched a division win, but had not clinched an outright win.

Therefore there was 0% chance Cleveland could beat them for the division during the regular season but a slight chance that the White Sox could still miss the playoffs.

IIRC

jabrch
06-11-2008, 05:05 PM
OK, their response to that may not have been very professional, but to completely discount a site because their playoff odds were slightly off due to some minor mathematical error is ridiculous.

Lip is not (I believe) only basing his position on one incident. I am sure that was icing on the cake.

The sheer arrogance of Sheehan and company that their ethos and math are correct and that anyone who disagrees is just not smart enough to understand is a major turn off.

I have taken plenty of statistics in both college and grad school. Lip's point of there being millions of variables that determine wins and losses is largely why I discount this discipline. Statistics are great to measure the outputs of a machine and predict future outputs. They are great for telling you what happened in complex environments. But when you try and use them to project how 50+ human beings, and many random or at least unpredictable events combine together and use that to make predictions is - as said above - guessing - with numbers behind it. That's not forecasting or analytics.

Frater Perdurabo
06-11-2008, 05:10 PM
They had clinched a division win, but had not clinched an outright win.

Therefore there was 0% chance Cleveland could beat them for the division during the regular season but a slight chance that the White Sox could still miss the playoffs.

IIRC

:?: :scratch:

Not picking on you, fquaye, but what is that supposed to mean? If they won the division, they make the playoffs. The second event (making the playoffs) happens 100% of the time when the first event happens (winning the division)! (Unless the team were to be swept away in a plane crash, asteroid impact or ebola outbreak.)

:scratch::?:

If BP or any other publication made a statement like that which I bolded above in your quote, they rightly should be ridiculed and be forced to sit in the corner, wear a dunce cap, eat paste, and hide under their pocket protectors.

Daver
06-11-2008, 05:11 PM
But when you try and use them to project how 50+ human beings, and many random or at least unpredictable events combine together and use that to make predictions is - as said above - guessing - with numbers behind it. That's not forecasting or analytics.

Exactly, it's mental masturbation, it's using flawed numbers to arrive at a conjectured conclusion, it's little more than a waste of time and energy.

fquaye149
06-11-2008, 05:12 PM
:?: :scratch:

Not picking on you, fquaye, but what is that supposed to mean? If they won the division, they make the playoffs. The second event (making the playoffs) happens 100% of the time when the first event happens (winning the division)! (Unless the team were to be swept away in a plane crash, asteroid impact or ebola outbreak.)

:scratch::?:


IIRC Cleveland could have tied us, forcing a playoff.

That means that although we would have finished the season in first place, we wouldn't have won the division outright, and then would have a 50% chance (maybe more or maybe less depending on whether Cleveland had a better or worse chancing of beating us) to win the division.

When you factor in WC, that gave us more ways to make the playoffs, but I think the point was that although the White Sox had clinched a 1st place finish they still conceivably could have missed the playoffs had they lost out and CLE won out

fquaye149
06-11-2008, 05:13 PM
Exactly, it's mental masturbation, it's using flawed numbers to arrive at a conjectured conclusion, it's little more than a waste of time and energy.

As opposed to talking about the White Sox on the internet....

Sports are sports. Unless you have a job in sports, it's all just mental masturbation :rolleyes:

doublem23
06-11-2008, 05:14 PM
They had clinched a division win, but had not clinched an outright win.

Therefore there was 0% chance Cleveland could beat them for the division during the regular season but a slight chance that the White Sox could still miss the playoffs.

IIRC

No, when the Sox "clinched" the division that afternoon in Detroit, they had not mathematically done so, had Cleveland swept the final series with the Sox, the Sox and Tribe would have been tied on the last day of the season. The reason the Sox could celebrate a division title is because that same weekend, the Yankees and Red Sox were playing a 3-game series in Boston, and because of that, if the Indians swept the Sox in Cleveland, both the Sox and Indians would finish with better records than whoever finished 2nd in the A.L. East, so the Sox were awarded the division before mathematically clinching it because if both the Sox and Indians qualified for the play-offs (as they would have to if the Indians swept the season's final series because of the NY/BOS series), because the Sox would still have a 10-9 egde in the season series, and MLB rules specifiy that if both teams would qualify for the post-season, then there is no play-off and the division is awarded on that tiebreaker.

The White Sox clinched the division in Detroit in 2005, not because they had mathematically done so, but because of circumstance. After they beat the Tigers, their "Magic Number" (a "stat" I believe is simple enough for everyone to understand) was still actually 1.

BP's projections probably showed the Sox's chances to win the division not at 100% because had the Yankees and Red Sox not been playing that same weekend, the White Sox would not have clinched the division after their game in Detroit, they would have had to win 1 more game or have Cleveland lose 1 more game. I guess you can chalk that up to laziness on someone's part for not updating their standings, but rather just letting the data run... :dunno:

Frater Perdurabo
06-11-2008, 05:15 PM
IIRC Cleveland could have tied us, forcing a playoff.

That means that although we would have finished the season in first place, we wouldn't have won the division outright, and then would have a 50% chance (maybe more or maybe less depending on whether Cleveland had a better or worse chancing of beating us) to win the division.

When you factor in WC, that gave us more ways to make the playoffs, but I think the point was that although the White Sox had clinched a 1st place finish they still conceivably could have missed the playoffs had they lost out and CLE won out

If you are tied for the division lead after Game 162, you have not won the division until the tiebreaker game(s), has (have) been played. If you lose the tiebreaker game(s), you don't win the division!

As a general rule, if you win the division, you are in the playoffs. Period.

Edit: What doublem said. And in the specific case of the 2005 AL Central, once the Sox "clinched" in Detroit, there was no way the Sox would not make the playoffs.

jabrch
06-11-2008, 05:16 PM
Exactly, it's mental masturbation, it's using flawed numbers to arrive at a conjectured conclusion, it's little more than a waste of time and energy.


Which, by the way, I don't mind... If that's your bag - cool.

But the claim that this is a science, a mathematical discipline, a way of making significant projections, just because it involves mathematics, is completely ridiculous. Add to that the arrogance that comes from the believers, and the "flat earth" characterizations they make of non-believers, and then adding on the follwers-on who don't even get the math behind the numbers - yet would follow it over the cliff like the dodo - and you get a really annoying crowd who I choose to ignore as much as possible.

I get the math behind it. But I disagree that the math is sensible or significant.

fquaye149
06-11-2008, 05:16 PM
If you are tied for the division lead after Game 162, you have not won the division until the tiebreaker game(s), has (have) been played. If you lose the tiebreaker game(s), you don't win the division!

If you win the division, you are in the playoffs. Period.

Ok......sure. :shrug:

Whatever the semantics, that was the situation, iirc

fquaye149
06-11-2008, 05:17 PM
No, when the Sox "clinched" the division that afternoon in Detroit, they had not mathematically done so, had Cleveland swept the final series with the Sox, the Sox and Tribe would have been tied on the last day of the season. The reason the Sox could celebrate a division title is because that same weekend, the Yankees and Red Sox were playing a 3-game series in Boston, and because of that, if the Indians swept the Sox in Cleveland, both the Sox and Indians would finish with better records than whoever finished 2nd in the A.L. East, so the Sox were awarded the division before mathematically clinching it because if both the Sox and Indians qualified for the play-offs (as they would have to if the Indians swept the season's final series because of the NY/BOS series), because the Sox would still have a 10-9 egde in the season series, and MLB rules specifiy that if both teams would qualify for the post-season, then there is no play-off and the division is awarded on that tiebreaker.

The White Sox clinched the division in Detroit in 2005, not because they had mathematically done so, but because of circumstance.

That's what I'm saying. I guess I didn't communicate it adeuqately

StepsInSC
06-11-2008, 05:18 PM
Lip:

Do you really think OBP is gobbledeegook? I can understand not wanting anything to do with stats that have no self-explanatory intrinsic value (OPS is the most commonly used one), but OBP is simply a measure of how much someone gets on base...it's as intuitive as batting average.

My only complaint in all this is that some people here always rail on "statheads" in these threads even when none of these statheads have posted in the thread. Without fail every single time BP is mentioned around here you'll get a rehashed one-liner from the same posters. We get it already. You hate them.

doublem23
06-11-2008, 05:22 PM
That's what I'm saying. I guess I didn't communicate it adeuqately

Right, but under the actual circumstances, after the White Sox beat the Tigers on Sep. 29, they had clinched the division and had qualified for the post-season, but it was only because the Yankees and Red Sox were playing that weekend, as well. If the Yankees and Sawx had closed out the season against any other opponents, that win over the Tigers would not have been a division clincher for the White Sox.

I think we're in the same page.

Frater Perdurabo
06-11-2008, 05:23 PM
Ok......sure. :shrug:

Whatever the semantics, that was the situation, iirc

Once the Sox won the game in Detroit and "clinched" the division, by circumstance/rule if not by math, they were going to the playoffs.

That's why it was completely asinine for BP or any other article to claim that the Sox still were not 100% certain to make the playoffs (other than a tragic accident of some kind).

BP put the math ahead of the actual circumstances. That's why they are justifiably ridiculed.

I'm not ridiculing you. I'm ridiculing those who value the math more than the actual results.

Iwritecode
06-11-2008, 05:27 PM
If you are tied for the division lead after Game 162, you have not won the division until the tiebreaker game(s), has (have) been played. If you lose the tiebreaker game(s), you don't win the division!

As a general rule, if you win the division, you are in the playoffs. Period.

Actually, they changed that a few years ago. If two teams in the same division are tied after 162 games, they use the H2H records to determine the division winner. Whatever team has more wins against the other one, wins the division.

It would be stupid to play one extra game to determine a division winner when both teams are going to the playoffs no matter what.

Frater Perdurabo
06-11-2008, 05:30 PM
Actually, they changed that a few years ago. If two teams in the same division are tied after 162 games, they use the H2H records to determine the division winner. Whatever team has more wins against the other one, wins the division.

It would be stupid to play one extra game to determine a division winner when both teams are going to the playoffs no matter what.

We're both right. I was speaking in the general sense; you're speaking in the specific instance of two teams in the same division being tied for the division lead, and also both teams having better records than any other 2nd place team in that league.
:smile:

jabrch
06-11-2008, 05:34 PM
Lip:

Do you really think OBP is gobbledeegook? I can understand not wanting anything to do with stats that have no self-explanatory intrinsic value (OPS is the most commonly used one), but OBP is simply a measure of how much someone gets on base...it's as intuitive as batting average.

My only complaint in all this is that some people here always rail on "statheads" in these threads even when none of these statheads have posted in the thread. Without fail every single time BP is mentioned around here you'll get a rehashed one-liner from the same posters. We get it already. You hate them.


I don't think OBP itself is gobbledegook. I think the people who take it to the point where they value a walk as much as a hit are gobbledegook.

OBP measures two different things.

Getting a Hit in an AB

and

Getting a Walk in a PA

The first is somewhat within the hitters control. The second is entirely withing the control of a pitcher. The first is correlated largely with a hitters' skill. The second is correlated much more with other factors besides how good a walker a guy is. The second is also highly dependent on the first. And just doing the second without the first is uncommon (save for a few guys who slug so much that pitchers "give" them walks more frequently) and often far less effective than the numbers might indicate.

Looking at OBP tells you very little without looking at he components of it - and understanding the type of hitter you are evaluating.

doublem23
06-11-2008, 05:37 PM
I don't think OBP itself is gobbledegook. I think the people who take it to the point where they value a walk as much as a hit are gobbledegook.

OBP measures two different things.

Getting a Hit in an AB

and

Getting a Walk in a PA

The first is somewhat within the hitters control. The second is entirely withing the control of a pitcher. The first is correlated largely with a hitters' skill. The second is correlated much more with other factors besides how good a walker a guy is. The second is also highly dependent on the first. And just doing the second without the first is uncommon (save for a few guys who slug so much that pitchers "give" them walks more frequently) and often far less effective than the numbers might indicate.

Looking at OBP tells you very little without looking at he components of it - and understanding the type of hitter you are evaluating.

I strongly beg to differ.

Eddo144
06-11-2008, 05:41 PM
I don't think OBP itself is gobbledegook. I think the people who take it to the point where they value a walk as much as a hit are gobbledegook.
First, anyone who values a walk as much as a hit is wrong. Even ask BP: they'll tell you a single is worth about 1.5 times as much as a walk.

OBP measures two different things.

Getting a Hit in an AB

and

Getting a Walk in a PA

The first is somewhat within the hitters control. The second is entirely withing the control of a pitcher. The first is correlated largely with a hitters' skill. The second is correlated much more with other factors besides how good a walker a guy is. The second is also highly dependent on the first. And just doing the second without the first is uncommon (save for a few guys who slug so much that pitchers "give" them walks more frequently) and often far less effective than the numbers might indicate.

Looking at OBP tells you very little without looking at he components of it - and understanding the type of hitter you are evaluating.
You really think hitters have no control over drawing walks? Then why do the same players lead the league year after year? And it's not just feared hitters; someone like Vlad Guerrero sure is feared, but he's never among the league leaders in walks.

Walking is a skill. Is it as good a skill to have as hitting or hitting for power? No. But it still has some value.

jabrch
06-11-2008, 05:42 PM
I strongly beg to differ.


A pitcher entirely controls if you CAN or CAN NOT walk. Then the execution is up to the player. A pitcher can (and many do) guarantee that you have no opportunity to walk.

Maybe my wording wasn't clear - but I am not sure what you would strongly differ with. (I know a hitter still has to take the pitch - but the pitcher is the one who throws a ball or throws a pitch that gives you ZERO chance to walk.

jabrch
06-11-2008, 05:45 PM
First, anyone who values a walk as much as a hit is wrong. Even ask BP: they'll tell you a single is worth about 1.5 times as much as a walk.

OBP values them EXACTLY the same. If you talk of OBP, you are using a metric that tells you each occurance of being on base is worth the same.


You really think hitters have no control over drawing walks?

I don't believe I said that... at least that wasn't the intent. Maybe my wording lacked precision.

I said that hitters can control their hitting. Pitchers control the pitch location that determines is a walk is possible. I know it is a skill. I know the hitter has a part in it. But a pitcher has 100% control in determining IF a batter will have the opportunity to walk - it is then up to that hitter to execute the walk...

jabrch
06-11-2008, 05:46 PM
:tomatoaward

doublem23
06-11-2008, 05:49 PM
OBP values them EXACTLY the same. If you talk of OBP, you are using a metric that tells you each occurance of being on base is worth the same.

A single is worth more because of the extra stuff that may come with it; defensive errors, chance for runners to move up an extra base, but a single with the bases empty or one that only allows baserunners to move up a single base in the end really isn't any more valuable than a walk, despite how much more difficult it may be to come by.

:dunno:

Eddo144
06-11-2008, 05:54 PM
OBP values them EXACTLY the same. If you talk of OBP, you are using a metric that tells you each occurance of being on base is worth the same.
Yes, you're right. That's why OBP isn't perfect.

But you know what, batting average is as bad, if not worse! Batting average gives a walk zero value.

OPS is an attempt to fix the relative worth issue, but it overvalues slugging. In OPS, roughly, a walk is worth 1, a single worth 2, a double worth 3, a triple worth 4, and a HR worth 5.

Something like EqA is better - I'm not sure of the exact calculation, but I do know that it values a single at 1.5 times a walk.

There's no perfect metric. You need to look at all of them. You definitely should be open to new ones - don't accept them at face value right away, but take a good look at what they measure. Don't call them "fiction" off the bat, until you actually learn something about them.

jabrch
06-11-2008, 05:55 PM
A single is worth more because of the extra stuff that may come with it; defensive errors, chance for runners to move up an extra base, but a single with the bases empty or one that only allows baserunners to move up a single base in the end really isn't any more valuable than a walk, despite how much more difficult it may be to come by.

:dunno:

What you said is true. But there is more than that. A hitter can get a hit on a perfect pitch. A hitter only walks on a series of four defects made by the pitcher.

doublem23
06-11-2008, 05:59 PM
What you said is true. But there is more than that. A hitter can get a hit on a perfect pitch. A hitter only walks on a series of four defects made by the pitcher.

Why do you assume that all balls are poor pitches? Many times, pitchers purposefully go out of the zone to try and get a hitter to chase a bad pitch. It's not like every pitcher is trying to throw the ball right down the middle of the plate and he only misses the strike zone by accident.

Also, a "perfect pitch" doesn't automatically mean it's a strike. Your telling me a guy who slaps a mistake over the plate by the pitcher is automatically more skilled than a guy who fouls off a number of pitches and then draws a walk after working the pitcher?

Eddo144
06-11-2008, 06:07 PM
OBP values them EXACTLY the same. If you talk of OBP, you are using a metric that tells you each occurance of being on base is worth the same.
To add to my previous reply, it's better to think of OBP as the inverse of what it's actually measuring - that is, out percentage.

OBP simply measures the percentage of the times you come up to the plate that you don't make an out.

Daver
06-11-2008, 06:14 PM
To add to my previous reply, it's better to think of OBP as the inverse of what it's actually measuring - that is, out percentage.

OBP simply measures the percentage of the times you come up to the plate that you don't make an out.

Well, that skews what OPS tells you then doesn't it?


See what I mean by mental masturbation?

jabrch
06-11-2008, 06:25 PM
Yes, you're right. That's why OBP isn't perfect.

But you know what, batting average is as bad, if not worse! Batting average gives a walk zero value.



and it doesn't profess to measure walks at all. It just measures hitting.

Now I wasn't arguing the merit of one or the other. In isolation, one tells you H/AB the other tells you OB/PA. Neither are worth much in terms of evaluating an overall hitter. But the more variables you cram into it, the less meaningful it becomes. That's why you can tell a lot more about a hitter from reading a raw stat sheet than you can from all the mental masturbation it takes to combine multiple items to make one statistic.

Again - my point was not to argue OBP vs AVG. My point was that a hitter controls hits and if a pitcher does what he wants to do, completely controls if a hitter has the opportunity to walk.

OPS is an attempt to fix the relative worth issue, but it overvalues slugging. In OPS, roughly, a walk is worth 1, a single worth 2, a double worth 3, a triple worth 4, and a HR worth 5.

Worse yet....
Adding X/A to Y/B does not equal X+Y/A+B. And you won't convince me otherwise. That's what OPS tries to do.

Something like EqA is better - I'm not sure of the exact calculation, but I do know that it values a single at 1.5 times a walk.

EQA is (H + TB + 1.5*(BB + HBP + SB) + SH + SF) divided by (AB + BB + HBP + SH + SF + CS + SB) then normalized to the league #. And at that point in time you have so much data into the number that it tells you nothing, overvalues certain types of hitters and undervalues others. The fact that a lot of players have high EQAs doesn't mean EQA means anything. It just means you took a lot of good numerators and divided them by a denominator.

There's no perfect metric.

We agree 100% on this.

You need to look at all of them.

Not really...I don't care about ones that are meaningless because they have too many variables to draw any conclusion.

but take a good look at what they measure. Don't call them "fiction" off the bat, until you actually learn something about them.

That's odd from the guy who uses EQA but doesn't know how it is calculated? They are fiction because they aren't mathematically significant tools to make projections just because you add/subtract/multiply/divide numbers.

anewman35
06-11-2008, 06:27 PM
Well, that skews what OPS tells you then doesn't it?


See what I mean by mental masturbation?

There is no one perfect baseball stat that tells you everything you need to know about a player. I don't think anybody would ever claim otherwise. That being said, I can't understand why it's somehow bad to try to get the best one possible.

jabrch
06-11-2008, 06:28 PM
OBP simply measures the percentage of the times you come up to the plate that you don't make an out.

First off - it isn't. Because an error doesn't help OBP and isn't an out. But disregard that for a second...

This is a horrendous way to look at baseball because all non-outs are not worth the same. And all outs are not worth the same.

What OBP is good for is to tell you how often a guy gets on base vs his # of opportunities. But that is nowhere near close to enough to evaluate a player. You have already stipulated to that. And it also DOES count a hit exactly the same as a walk, which is clearly flawed.

anewman35
06-11-2008, 06:29 PM
I'm late to this all, but all I really have to say is that when one side resorts to childish insults ("propellerheads"), that's pretty much always the side that's wrong. And Lip complained a few pages back that BP was rude to people who complained in 2005. That may or may not be true, but I'd bet anything that most of the letters they got were highly rude (see, again "propellerheads", etc), so I don't really blame them.

jabrch
06-11-2008, 06:31 PM
There is no one perfect baseball stat that tells you everything you need to know about a player. I don't think anybody would ever claim otherwise. That being said, I can't understand why it's somehow bad to try to get the best one possible.

There is no best ONE possible. It is best to watch the games, read the box scores, look at the overall stat sheets and understand what happened, not try to simplify it to one number to satisfy a generation of media and sports fans who are so short sighted that they need a cut and dried better and best and can't handle that evaluations such as this are highly subjective and not at all quantifiable when aggregated at the overall level of evaluating "hitting".

jabrch
06-11-2008, 06:32 PM
I'm late to this all, but all I really have to say is that when one side resorts to childish insults ("propellerheads"), that's pretty much always the side that's wrong. And Lip complained a few pages back that BP was rude to people who complained in 2005. That may or may not be true, but I'd bet anything that most of the letters they got were highly rude (see, again "propellerheads", etc), so I don't really blame them.


Right and wrong can't be determined by # of childish insults...puhleez.

anewman35
06-11-2008, 06:36 PM
Right and wrong can't be determined by # of childish insults...puhleez.

No, but they don't help make your case. I feel the same way when other Sox fans talk about the "Scrubs" or whatever. It's just my personal opinion, but if a person uses terms like that, I pretty much disregard anything they have to say. I doubt I'm the only one. There's just pretty much never a good reason to bring a conversation down to the level of name calling.

jabrch
06-11-2008, 06:40 PM
No, but they don't help make your case. I feel the same way when other Sox fans talk about the "Scrubs" or whatever. It's just my personal opinion, but if a person uses terms like that, I pretty much disregard anything they have to say. I doubt I'm the only one. There's just pretty much never a good reason to bring a conversation down to the level of name calling.

I agree with that 100%. But to say that, "when one side resorts to childish insults ("propellerheads"), that's pretty much always the side that's wrong" is misleading at best. It certainly isn't logically congruent to the position that we should use more complicated statistics to make evaluations. In fact - it goes entirely the other direction.

Daver
06-11-2008, 06:47 PM
For the record, the term propellerhead was first used by statisticians to define themselves, it is not an insult, it is a self given title.

Lip Man 1
06-11-2008, 06:47 PM
Two things:

1. I hope someone can find that thread here at WSI because it either linked to the actual article or posted it directly.

2. I understand the nuance difference with what Double and some were saying about "clinching" and this is my fault for not being more specific. AFTER the Sox had 100% clinched by beating Cleveland that Friday they STILL did not change the percentage figure (if memory serves) and they didn't for awhile after that! That was why they were called out for it directly by fans and that's where the moron decided to attack Sox fans at their site.

Anewman I'm not making this up. My memory is clear on reading what that site said about Sox fans, here at WSI. What I'm not sure of, if it was Sheehan and his BP site. But I am sure of a thread here at WSI.

Lip

anewman35
06-11-2008, 06:50 PM
Anewman I'm not making this up. My memory is clear on reading what that site said about Sox fans, here at WSI. What I'm not sure of, if it was Sheehan and his BP site. But I am sure of a thread here at WSI.

Lip

I remember that too, I was here. I'm just saying, I suspect that if they were getting lots of complaints about it, many of them weren't especially polite. I see how many people here treat anybody who mentions stats.

voodoochile
06-11-2008, 06:51 PM
OPS is nothing more than a meta stat. It won't tell you anything about what a guy will do in a given at bat. It can only be used to compare players. Since the HOF is loaded with hitters with an OPS of .950 and up it might give you a simple statistical way to evaluate whether a given hitter is a HOF candidate, but it's not the be all and end all because all it does is show offensive production.

Years ago on another version of this board I proposed a stat: Total Bases Generated (TBG). Which evaluates a player based on how many total bases are generated on average for every PA.

Walk with first base open = 1 TBG
Single with men on second and third where both men score = 4 TBG (1+2+1)

Obviously the pinnacle of a single PA would be a Grand Slam which would be 10 TBG for that AB.

I'd like to see somone actually try to track that stat, but wouldn't be surprised if it came out around OPS. When I ran a guestimate from Frank's stats at the time I proposed it, it came out pretty close to his OPS that year.

Still, OPS isn't a reliable stat other than to compare players. It means nothing in terms of predictive ability.

Eddo144
06-11-2008, 07:06 PM
That's odd from the guy who uses EQA but doesn't know how it is calculated? They are fiction because they aren't mathematically significant tools to make projections just because you add/subtract/multiply/divide numbers.
OK, I've seen the EqA formula multiple times. I got lazy and didn't want to look it up in the middle of posting. I should be more diligent next time, I suppose.

And secondly, then everything is fiction. Don't listen to your weatherman - he's just adding and subtracting numbers. Don't listen to your financial adviser - anything he tells you is plain fiction.

Come on.

Eddo144
06-11-2008, 07:07 PM
First off - it isn't. Because an error doesn't help OBP and isn't an out. But disregard that for a second...

This is a horrendous way to look at baseball because all non-outs are not worth the same. And all outs are not worth the same.

What OBP is good for is to tell you how often a guy gets on base vs his # of opportunities. But that is nowhere near close to enough to evaluate a player. You have already stipulated to that. And it also DOES count a hit exactly the same as a walk, which is clearly flawed.
No kidding. There's a weakness of BA as well. And RBI. And hits. And runs. And everything.

Look, you can find something wrong with any statistic. Or any measure of any kind, even observation ("You saw it, but I didn't", for example).

Daver
06-11-2008, 07:23 PM
OK, I've seen the EqA formula multiple times. I got lazy and didn't want to look it up in the middle of posting. I should be more diligent next time, I suppose.

And secondly, then everything is fiction. Don't listen to your weatherman - he's just adding and subtracting numbers. Don't listen to your financial adviser - anything he tells you is plain fiction.

Come on.

The weatherman is the only job I know of where you can be dead wrong on a regular basis and still get paid.

As for finances, last time I checked if you add a dollar to a couple of quarters, it is going to add up to $1.50 every damn time you count it, there is no variable involved.

fquaye149
06-11-2008, 07:24 PM
Once the Sox won the game in Detroit and "clinched" the division, by circumstance/rule if not by math, they were going to the playoffs.

That's why it was completely asinine for BP or any other article to claim that the Sox still were not 100% certain to make the playoffs (other than a tragic accident of some kind).

BP put the math ahead of the actual circumstances. That's why they are justifiably ridiculed.

I'm not ridiculing you. I'm ridiculing those who value the math more than the actual results.

The claim they made is that the Sox were not 100% to win their division because even though they were 100% to go to the playoffs, they might have gone as wildcard

Eddo144
06-11-2008, 07:44 PM
As for finances, last time I checked if you add a dollar to a couple of quarters, it is going to add up to $1.50 every damn time you count it, there is no variable involved.
Same for baseball. If I go 1 for 3 with a walk today, my OBP is .500. If I go 1 for 3 with walk tomorrow, my OBP is .500.

Come on, Daver, now you're arguing that any math related to baseball is automatically wrong and worthless?

Daver
06-11-2008, 08:21 PM
Same for baseball. If I go 1 for 3 with a walk today, my OBP is .500. If I go 1 for 3 with walk tomorrow, my OBP is .500.

Come on, Daver, now you're arguing that any math related to baseball is automatically wrong and worthless?

When there is a variable like human decision involved, your numbers become worthless, fuzzy math is what it is, it's an exercise in mental masturbation. Most stats beyond the very basic ones exist for no other purpose than to add bargaining points to agents, nothing more and nothing less. There is a direct correlation between the advent of free agency and the sudden proliferation of stats that track useless crap like holds. Agents, and agents alone created the the role of the closer as it exists today, even though the save remains one of the most worthless stats in baseball.

Stats serve a very small purpose, they are a record of the past, the lettering on the tombstone of a dead season so to speak, beyond that they do little else.

Elephant
06-11-2008, 08:35 PM
IIRC, when the Sox clinched the division and BP gave them an 94% chance of making the playoffs, that was because there was a slight chance the Sox would get eaten by a bear.

ARTICLE IX RULE 1: A. No bear stool allowed to play baseball. B. Also no blacks.

I think B was repealed in the 40s.

Eddo144
06-11-2008, 08:35 PM
When there is a variable like human decision involved, your numbers become worthless, fuzzy math is what it is, it's an exercise in mental masturbation. Most stats beyond the very basic ones exist for no other purpose than to add bargaining points to agents, nothing more and nothing less. There is a direct correlation between the advent of free agency and the sudden proliferation of stats that track useless crap like holds. Agents, and agents alone created the the role of the closer as it exists today, even though the save remains one of the most worthless stats in baseball.

Stats serve a very small purpose, they are a record of the past, the lettering on the tombstone of a dead season so to speak, beyond that they do little else.
Wow. I mean, I can't really dispute much of what you're saying (though I still don't quite buy your agents-created-most-stats theory; do you have any articles or anything to back that up?). Still, if your true belief is that the record of the past is so unimportant, I guess you and I are just two very different people.

I mean, if that's the case, why hang pennants from previous years? Or statues of franchise greats? They're just records of the past, lettering on the tombstones of dead seasons.

Daver
06-11-2008, 08:54 PM
Wow. I mean, I can't really dispute much of what you're saying (though I still don't quite buy your agents-created-most-stats theory; do you have any articles or anything to back that up?). Still, if your true belief is that the record of the past is so unimportant, I guess you and I are just two very different people.

I mean, if that's the case, why hang pennants from previous years? Or statues of franchise greats? They're just records of the past, lettering on the tombstones of dead seasons.

Baseball FA began in 1976, as noted by Charlie Finley selling the A's as opposed to embracing the new concept, do your own research on what stats were tracked pre 1976, you will find they are pretty much limited to what is on the back of a baseball card. Tell me when STATS inc. came into existence, they created most of the useless stats we have now, their creation was funded by a consortium of player agents that were willing to pay for their service.

Arguing with the truth is a difficult task.

jabrch
06-11-2008, 09:01 PM
OK, I've seen the EqA formula multiple times. I got lazy and didn't want to look it up in the middle of posting. I should be more diligent next time, I suppose.

And secondly, then everything is fiction. Don't listen to your weatherman - he's just adding and subtracting numbers. Don't listen to your financial adviser - anything he tells you is plain fiction.

Come on.


Weathermen - if you want to question the science behind meteorology, that's your choice. Don't lump me in with you - you are wrong. That said, as Daver pointed out, since thier success ratio is not 6 sigma, one could question that science and not really be a complete "flat earther".

As far as my financial advisor, I'm not convinced he's much a scientist. He follows trends, does research, understands his market, but I don't believe there is a ton of mathematical rigour behind his work. (I'm talking my advisor, not a wall street analyst) When the market does well, my guy does well. When the market sucks, my guy kinda sucks. So maybe I need a new guy? Either way, he's not professing that his way of picking stocks is right and the way Warren Buffet has done it for 50 years was dead wrong - so I don't really care.

johnr1note
06-11-2008, 09:46 PM
I am not one to ever defend BP's propeller heads, but one of their kind, Nate Silver, wrote an article (http://www.baseballprospectus.com/unfiltered/?p=792) about how particular teams tend to beat their Pecota projections. Interestingly, the White Sox were one of the top 3 teams that beat BP's PEcota projections, while the Cubs were one of the worst. Mr. Silver refers to this as the "Most heart" factor.

In commenting on teams like the Braves, Angels, and White Sox, who have all regularly played much better than the BP Projections, Silver stated:

"Getting (modestly) more serious for a moment, there may be some systematic factors in the teams that tend to beat their PECOTA projections. The teams that have finished ahead of PECOTA tend to be well-run, cohesive organizations with good scouting departments and organizational cultures; these teams may have more capacity to right the ship as the season advances onward."

I take this as an admission on his part that Pecota is an imperfect system for predicting baseball success.

Elephant
06-11-2008, 11:19 PM
I am not one to ever defend BP's propeller heads, but one of their kind, Nate Silver, wrote an article (http://www.baseballprospectus.com/unfiltered/?p=792) about how particular teams tend to beat their Pecota projections. Interestingly, the White Sox were one of the top 3 teams that beat BP's PEcota projections, while the Cubs were one of the worst. Mr. Silver refers to this as the "Most heart" factor.

In commenting on teams like the Braves, Angels, and White Sox, who have all regularly played much better than the BP Projections, Silver stated:

"Getting (modestly) more serious for a moment, there may be some systematic factors in the teams that tend to beat their PECOTA projections. The teams that have finished ahead of PECOTA tend to be well-run, cohesive organizations with good scouting departments and organizational cultures; these teams may have more capacity to right the ship as the season advances onward."I take this as an admission on his part that Pecota is an imperfect system for predicting baseball success.

http://www.cracked.com/blog/wp-content/uploads/2007/12/o_rly.jpg


So you're saying stats don't tell the whole tale huh?

Stats are awesome, and they're very useful--especially OPS+. Either understand that or don't understand baseball. I always thought Roberto Alomar was a better hitter than Craig Biggio, and OPS+ confirms that. RF helps confirm that Fox was better than Sandberg, both multiple gold glove winners (and Sandberg won about three times as many, naturally, because he had power). Uribe has a higher RF than Cabrera, just like I would've expected. Almost all of them are useful in some way for understanding the game better. That's why they exist.

But like all things, some people champion them way too much and people who don't really understand them trash them categorically. Then there are smart people in the middle who take them for what they're worth.

FedEx227
06-11-2008, 11:24 PM
Baseball FA began in 1976, as noted by Charlie Finley selling the A's as opposed to embracing the new concept, do your own research on what stats were tracked pre 1976, you will find they are pretty much limited to what is on the back of a baseball card. Tell me when STATS inc. came into existence, they created most of the useless stats we have now, their creation was funded by a consortium of player agents that were willing to pay for their service.

Arguing with the truth is a difficult task.

Yeah because in 1954 Branch Rickey definitely did not release a pamphlet along with the Dodgers personal statistician Allan Roth about the powers of statistical research in baseball, as well as the notions that on-base percentage was a very superior stat to batting average and that platoon effects are real among other things detailed in this piece.

No way. Never happened.

FedEx227
06-11-2008, 11:25 PM
I am not one to ever defend BP's propeller heads, but one of their kind, Nate Silver, wrote an article (http://www.baseballprospectus.com/unfiltered/?p=792) about how particular teams tend to beat their Pecota projections. Interestingly, the White Sox were one of the top 3 teams that beat BP's PEcota projections, while the Cubs were one of the worst. Mr. Silver refers to this as the "Most heart" factor.

In commenting on teams like the Braves, Angels, and White Sox, who have all regularly played much better than the BP Projections, Silver stated:

"Getting (modestly) more serious for a moment, there may be some systematic factors in the teams that tend to beat their PECOTA projections. The teams that have finished ahead of PECOTA tend to be well-run, cohesive organizations with good scouting departments and organizational cultures; these teams may have more capacity to right the ship as the season advances onward."

I take this as an admission on his part that Pecota is an imperfect system for predicting baseball success.

No, that never exists. They never admit defeat, the never treat it as a mere prediction that means nothing in the grand scheme of baseball. No, absolutely not. Get out of here with that.

Lefty34
06-11-2008, 11:33 PM
When there is a variable like human decision involved, your numbers become worthless, fuzzy math is what it is, it's an exercise in mental masturbation. Most stats beyond the very basic ones exist for no other purpose than to add bargaining points to agents, nothing more and nothing less. There is a direct correlation between the advent of free agency and the sudden proliferation of stats that track useless crap like holds. Agents, and agents alone created the the role of the closer as it exists today, even though the save remains one of the most worthless stats in baseball.

Stats serve a very small purpose, they are a record of the past, the lettering on the tombstone of a dead season so to speak, beyond that they do little else.

Right. I always thought OPS+ and ERA+ were meant to show how a player was performing against the league average, thus differentiating between a hitter that OPSes .800 in a year when power numbers were way up and one who OPSes at the same level when offensive numbers are down across the league. Thanks for setting me straight.

FedEx227
06-11-2008, 11:35 PM
Right. I always thought OPS+ and ERA+ were meant to show how a pitcher was performing against the league average, thus differentiating between a hitter that OPSes .800 in a year when power numbers were way up and one who OPSes at the same level when offensive numbers are down across the league. Thanks for setting me straight.

No stupid. You're a moron, just go finish off the last batch of Teddy Grahams and ask your mom to get some from the store tomorrow.

Lefty34
06-11-2008, 11:41 PM
It saves a lot of typing too.

I can watch someone play baseball and have a good idea whether they are good or bad at it.

Strange concept to the proppelerheads, but it has worked for over a hundred years.

You're right, it has worked (for some players, others, not so much). Oh wait, there were those 40+ years of baseball history where you were immediately passed over by scouts' eyes if you were black. Oh well, numbers suck. Bad bad bad.

Elephant
06-11-2008, 11:47 PM
Stats are only the primary measure of how we as fans judge all players in all sports, except maybe offensive linemen.

If you think otherwise, you're full of ****. I never watched Jeff Bagwell much, but I can tell you he's right up there with Frank, whom I watched for over a decade. Now would I know Bagwell was great without stats? Sure. But would I be able to prove that he put up practically identical numbers to him in the same era, and is among the elite all time first baseman? No.

DSpivack
06-11-2008, 11:58 PM
Stats are only the primary measure of how we as fans judge all players in all sports, except maybe offensive linemen.

If you think otherwise, you're full of ****. I never watched Jeff Bagwell much, but I can tell you he's right up there with Frank, whom I watched for over a decade. Now would I know Bagwell was great without stats? Sure. But would I be able to prove that he put up practically identical numbers to him in the same era, and is among the elite all time first baseman? No.

Baseball, for whatever reason, is the most stat-heavy sport. In soccer, the world's most popular sport that most Americans don't generally care for, stats are mostly ignored.

FedEx227
06-12-2008, 12:13 AM
Baseball, for whatever reason, is the most stat-heavy sport. In soccer, the world's most popular sport that most Americans don't generally care for, stats are mostly ignored.

This has been asked a lot why baseball is so stat-heavy while others are not and I think it depends a lot on how the game is played. It's typically a one on one sport for the most part, but so many factors can potentially go into who wins the one on one deal. Pitch counts, score, runners on base, defensive miscues, etc.

You don't get that with another sport. In soccer it's not as if a player will try to score a goal in a different way if the score is 4-3 or 4-0. Also, in baseball one player (offensively) does not have as much impact on his team as any other sport, thus you better be damn certain the guys around him are also skilled. In soccer or in basketball a superstar can carry your team. Albert Pujols cannot single-handedly carries the Cardinals to a successful season, it's impossible. He can put up monster numbers, play great defense, but it still wouldn't mean a thing if a bunch of scrubs are filling out the rest of the positions.

Which relates to another baseball nuiance, in the offense can be taken away from a "superstar". You can't all of a sudden not make Kobe Bryant get the ball, he'll get it eventually. In baseball, if Albert Pujols is the one decent player on a team filled with Single A rejects, then he'll be walked and those Single A rejects are now responsible for scoring him.

In basketball if it's a key situation or you need points fast, you go to your star... ditto for soccer. In baseball you have no control over that, you have to work with who is up, thus you end of getting a lot of research and talk to ensure that the best possible players are out there with regards to budget, etc.

DSpivack
06-12-2008, 12:40 AM
This has been asked a lot why baseball is so stat-heavy while others are not and I think it depends a lot on how the game is played. It's typically a one on one sport for the most part, but so many factors can potentially go into who wins the one on one deal. Pitch counts, score, runners on base, defensive miscues, etc.

You don't get that with another sport. In soccer it's not as if a player will try to score a goal in a different way if the score is 4-3 or 4-0. Also, in baseball one player (offensively) does not have as much impact on his team as any other sport, thus you better be damn certain the guys around him are also skilled. In soccer or in basketball a superstar can carry your team. Albert Pujols cannot single-handedly carries the Cardinals to a successful season, it's impossible. He can put up monster numbers, play great defense, but it still wouldn't mean a thing if a bunch of scrubs are filling out the rest of the positions.

Which relates to another baseball nuiance, in the offense can be taken away from a "superstar". You can't all of a sudden not make Kobe Bryant get the ball, he'll get it eventually. In baseball, if Albert Pujols is the one decent player on a team filled with Single A rejects, then he'll be walked and those Single A rejects are now responsible for scoring him.

In basketball if it's a key situation or you need points fast, you go to your star... ditto for soccer. In baseball you have no control over that, you have to work with who is up, thus you end of getting a lot of research and talk to ensure that the best possible players are out there with regards to budget, etc.

This is getting away from the stats discussion on the thread, but baseball also is a completely unique game. Basketball, soccer, hockey, football, rugby and others all are based around a similar idea; get a ball or puck from one end of a field/court/rink to the other, into a goal/net/hoop/endzone, with the opposing side trying to stop you. An equal on either side, be it 5 or 11 or whatever.

While one player can not carry a team, it is in many ways a very individualistic sport. I think it is a great American sport because it is so indivualized, the drama is heightened when it's a showdown between pitcher and batter, one trying to stop the other.

FedEx227
06-12-2008, 12:46 AM
This is getting away from the stats discussion on the thread, but baseball also is a completely unique game. Basketball, soccer, hockey, football, rugby and others all are based around a similar idea; get a ball or puck from one end of a field/court/rink to the other, into a goal/net/hoop/endzone, with the opposing side trying to stop you. An equal on either side, be it 5 or 11 or whatever.

While one player can not carry a team, it is in many ways a very individualistic sport. I think it is a great American sport because it is so indivualized, the drama is heightened when it's a showdown between pitcher and batter, one trying to stop the other.

Absolutely, it's really in the bare essence a one on one showdown which is unique amongst most major sports.

fquaye149
06-12-2008, 05:43 AM
You're right, it has worked (for some players, others, not so much). Oh wait, there were those 40+ years of baseball history where you were immediately passed over by scouts' eyes if you were black. Oh well, numbers suck. Bad bad bad.

That has absolutely nothing to do with the task at hand.

It's not like baseball would have been desegregated if you could have convinced scouts through numbers that black players were good.

Everyone knew black players were good. The league simply had a "gentleman's" agreement that no team would sign one.

Thank God for Branch Rickey.

fquaye149
06-12-2008, 05:45 AM
Baseball, for whatever reason, is the most stat-heavy sport. In soccer, the world's most popular sport that most Americans don't generally care for, stats are mostly ignored.

That's because

a.) there aren't an abundance of stat-producing situations in soccer

b.) europeans call it football and that opens up a whole new can of worms. Think about trying to calculate Thierry Henry's QB Rating and you will realize why soccer has no stats. It simply can not be done.

c.) In England they call elevators "lifts"

d.) Is it true French Babes don't shave their pits?

Ava83
06-12-2008, 09:56 AM
A girl asks a simple question! WOW...Lots of interesting reading here.

Optipessimism
06-12-2008, 11:15 AM
A girl asks a simple question! WOW...Lots of interesting reading here.
Girls always have to start trouble.

doublem23
06-12-2008, 11:17 AM
Stats are only the primary measure of how we as fans judge all players in all sports, except maybe offensive linemen.

Not surprisingly, offensive lineman is the most misunderstood/underappreciated position in all of professional sports.

fquaye149
06-12-2008, 11:43 AM
Not surprisingly, offensive lineman is the most misunderstood/underappreciated position in all of professional sports.

Jonathan Ogden is a hall of famer!

Ava83
06-12-2008, 11:54 AM
Girls always have to start trouble.

(before I answer let me freshen up my lipstick and bat my eyes)

Girls NEVER intentionally start trouble!

(to be honest I had no clue it would go on for so long)

Lefty34
06-12-2008, 08:04 PM
That has absolutely nothing to do with the task at hand.

It's not like baseball would have been desegregated if you could have convinced scouts through numbers that black players were good.

Everyone knew black players were good. The league simply had a "gentleman's" agreement that no team would sign one.

Thank God for Branch Rickey.

Not really. Daver was saying (or at least implying) that numbers are meaningless when scouting MLB talent, and that that scouting has worked for the last century. In fact, many black players were far superior to their white counterparts in the MLB at that time, and their numbers showed it (though they were numbers from a separate league, they were still phenomenal). The numbers showed scouts that black players were very good yet they still refused to sign them based on their ignorance and bias, and therein lies the problem: scouts can be just as biased (if not more so) than numbers ever could be, and the decades of segregation in the MLB prove that.

Especially in the Stats v. Non-Stats discussions here, many people take the Non-Stats (scouting) side because "that's the way it has been done for a long time, and numbers cannot do the job of a scout any better because the game isn't played on paper". However up until the late 40's, many statistically (and actually) better ballplayers were not scouted and/or not signed by major league scouts simply because they were black. Isn't that the very definition of the bias that people claim numbers have? Someone can cherry-pick stats to make a player seem better than he is, but that should not be grounds for rendering statistics meaningless and scouting the Holy Grail of personnel decisions, especially when we have such a clear-cut example of scout bias staring at us in the face.

getonbckthr
06-12-2008, 08:12 PM
Maybe they should use the "Pick-To-Click" scoring system to determine a players value.

Daver
06-12-2008, 08:13 PM
Not really. Daver was saying (or at least implying) that numbers are meaningless when scouting MLB talent, and that that scouting has worked for the last century. In fact, many black players were far superior to their white counterparts in the MLB at that time, and their numbers showed it (though they were numbers from a separate league, they were still phenomenal). The numbers showed scouts that black players were very good yet they still refused to sign them based on their ignorance and bias, and therein lies the problem: scouts can be just as biased (if not more so) than numbers ever could be, and the decades of segregation in the MLB prove that.

Especially in the Stats v. Non-Stats discussions here, many people take the Non-Stats (scouting) side because "that's the way it has been done for a long time, and numbers cannot do the job of a scout any better because the game isn't played on paper". However up until the late 40's, many statistically (and actually) better ballplayers were not scouted and/or not signed by major league scouts simply because they were black. Isn't that the very definition of the bias that people claim numbers have? Someone can cherry-pick stats to make a player seem better than he is, but that should not be grounds for rendering statistics meaningless and scouting the Holy Grail of personnel decisions, especially when we have such a clear-cut example of scout bias staring at us in the face.

Any number that is dependent on human decision is a flawed number, your stats are based on fuzzy math. This is a fact that you can't change, and none of the numbers can either. If you can't grasp that simple dynamic, I really don't know what to tell you.


And for whoever the moron that brought up the black ballplayers that were overlooked, your point, if you even had one, is lost in your stupidity for not knowing your history.

Lefty34
06-12-2008, 08:28 PM
Any number that is dependent on human decision is a flawed number, your stats are based on fuzzy math. This is a fact that you can't change, and none of the numbers can either. If you can't grasp that simple dynamic, I really don't know what to tell you.


What dynamic? What are you talking about? AVG is a number based on human decision (swing or not swing), and it sucks in comparison to other stats like OBP (swing, no swing and ability to otherwise get on base). Yes, OBP still takes into account human decisions, but it is still far superior in evaluating baseball talent. OPS combines SLG with OBP, even further distancing itself from the "human decisions" you speak of, and OPS+ rates that distanced stat against the league. You can't arbitrarily say stats suck just because you think they are based on some phantom, all-knowing "human decision". Players just can't up and decide to start hitting for power and hit balls into the gap, they can say they want to do it, but that gets them nowhere near actually doing it.

Daver
06-12-2008, 08:40 PM
What dynamic? What are you talking about? AVG is a number based on human decision (swing or not swing), and it sucks in comparison to other stats like OBP (swing, no swing and ability to otherwise get on base). Yes, OBP still takes into account human decisions, but it is still far superior in evaluating baseball talent. OPS combines SLG with OBP, even further distancing itself from the "human decisions" you speak of, and OPS+ rates that distanced stat against the league. You can't arbitrarily say stats suck just because you think they are based on some phantom, all-knowing "human decision". Players just can't up and decide to start hitting for power and hit balls into the gap, they can say they want to do it, but that gets them nowhere near actually doing it.

Is every pitch called by a human umpire?

Yes.

Is every play on the field decided by a human umpire?

Yes

Do humans make errors that can skew numbers one way or the other?

Yes.

Are the numbers that track this skewed because of human decisions?

Yes.

The logic isn't hard, I would have thought a propellerhead could grasp it.

getonbckthr
06-12-2008, 08:40 PM
What dynamic? What are you talking about? AVG is a number based on human decision (swing or not swing), and it sucks in comparison to other stats like OBP (swing, no swing and ability to otherwise get on base). Yes, OBP still takes into account human decisions, but it is still far superior in evaluating baseball talent. OPS combines SLG with OBP, even further distancing itself from the "human decisions" you speak of, and OPS+ rates that distanced stat against the league. You can't arbitrarily say stats suck just because you think they are based on some phantom, all-knowing "human decision". Players just can't up and decide to start hitting for power and hit balls into the gap, they can say they want to do it, but that gets them nowhere near actually doing it.
So your batting average increases everytime you swing? If anything average is a number based upon the number of at bats and how often, on average, you reach base.

FedEx227
06-12-2008, 08:45 PM
So your batting average increases everytime you swing? If anything average is a number based upon the number of at bats and how often, on average, you reach base.

Reach base VIA hit. Only hit. Walks are BAD!

getonbckthr
06-12-2008, 08:53 PM
Reach base VIA hit. Only hit. Walks are BAD!
Ya I know I was just trying to show how stupid his post sounded. Sorry about the confusion.

FedEx227
06-12-2008, 08:57 PM
Ya I know I was just trying to show how stupid his post sounded. Sorry about the confusion.

But there in lies the problem with batting average that old times for whatever reason still can't stand. Great, I understand Chadwick, who wouldn't know a game of baseball if it bit him on ass, decides this is how it should be scored. And then they put it on baseball cards and old timers REFUSE to let it change.

Batting average is worthless. It's just as dependent on human error as OBP and leaves out an absolutely vital part of baseball.

getonbckthr
06-12-2008, 09:03 PM
But there in lies the problem with batting average that old times for whatever reason still can't stand. Great, I understand Chadwick, who wouldn't know a game of baseball if it bit him on ass, decides this is how it should be scored. And then they put it on baseball cards and old timers REFUSE to let it change.

Batting average is worthless. It's just as dependent on human error as OBP and leaves out an absolutely vital part of baseball.
I always saw it as such:
Batting average: percentage of successful attempts to bat (hit) and reach base.
Walking average: amount of times successful reaching bases without batting (hitting) a ball.
On Base %= B Ave+ W Ave.
I know that sounds really retarded.:tongue:

Lefty34
06-12-2008, 09:19 PM
So your batting average increases everytime you swing? If anything average is a number based upon the number of at bats and how often, on average, you reach base.

Actually no, what you just decribed would be OBP, not AVG. Like FedEx has said, AVG is worthless when compared to OBP.

And yes my post did sound a little dumb, I was just so blinded with rage that I couldn't get the words to form anything resembling a coherent thought.

doublem23
06-12-2008, 09:24 PM
Batting average is worthless now? :scratch: You should be Juan Uribe's agent. I thought we all already agreed that a single > walk.

Daver
06-12-2008, 09:24 PM
Actually no, what you just decribed would be OBP, not AVG. Like FedEx has said, AVG is worthless when compared to OBP.

And yes my post did sound a little dumb, I was just so blinded with rage that I couldn't get the words to form anything resembling a coherent thought.

The truth hurts.

Lefty34
06-12-2008, 09:27 PM
Batting average is worthless now? :scratch: You should be Juan Uribe's agent. I thought we all already agreed that a single > walk.

No, you are taking what I said out of context. I said that AVG is worthless when compared with OBP, because AVG does not take into account BB's, and it doesn't even count them as AB's. In terms of the worth of a single versus a walk, yes a single has more worth in terms of average runs produced (I believe), but this does not make AVG a better stat, it is actually a very incomplete stat.

voodoochile
06-12-2008, 09:39 PM
But there in lies the problem with batting average that old times for whatever reason still can't stand. Great, I understand Chadwick, who wouldn't know a game of baseball if it bit him on ass, decides this is how it should be scored. And then they put it on baseball cards and old timers REFUSE to let it change.

Batting average is worthless. It's just as dependent on human error as OBP and leaves out an absolutely vital part of baseball.

Actually no, what you just decribed would be OBP, not AVG. Like FedEx has said, AVG is worthless when compared to OBP.

And yes my post did sound a little dumb, I was just so blinded with rage that I couldn't get the words to form anything resembling a coherent thought.

And you wonder why people think statheads are blind to the realities of baseball?

What would you rather have, a player with a .320 avg/ .380 OBP

or

A player with a .270 avg and a .380 OBP?

If you chose the second player, I think you're an idiot. That simple test shows that avg is not a worthless stat.