PDA

View Full Version : Relief Pitchers and "Inherited Runners Scored"


chisoxfanatic
09-03-2008, 09:10 PM
I've been thinking of this a while, actually. ERA, WHIP, record, holds, and saves are pretty standard statistics for relief pitchers. You'll see all of those numbers in the stat box whenever a relief pitcher enters the game. I, however, never have seen them list anything about how many inherited runners they allow to score, which I think is a very telling stat in determining the quality of a relief pitcher (they're paid to not only get their guys out, but to strand runners as well). You'll rarely even hear the commentators discuss it either. I'm wondering why not as much of a big deal is made out of this. It should be one of the major relief pitching stats, or at least something that deals with it.

turners56
09-03-2008, 10:09 PM
I've been thinking. Maybe there should be a stat called reliever ERA. It would be abbreviated as rERA. It would count all earned runs, but it will also count inherited runners scored with the exception of inherited runners on third. Anybody like it?

chisoxfanatic
09-03-2008, 10:13 PM
I've been thinking. Maybe there should be a stat called reliever ERA. It would be abbreviated as rERA. It would count all earned runs, but it will also count inherited runners scored with the exception of inherited runners on third. Anybody like it?
We could always have an IRS in baseball. :cool:

pierzynski07
09-03-2008, 10:13 PM
I've been thinking. Maybe there should be a stat called reliever ERA. It would be abbreviated as rERA. It would count all earned runs, but it will also count inherited runners scored with the exception of inherited runners on third. Anybody like it?
Nope.

I've seen the inherited runner stat used quite a bit in broadcasts, so it is being kept track of.

chisoxfanatic
09-03-2008, 10:17 PM
Nope.

I've seen the inherited runner stat used quite a bit in broadcasts, so it is being kept track of.
It would be nice for it to be easier to find like the other "major" statistics. They should put it in that little stat box.

DumpJerry
09-03-2008, 10:27 PM
I prefer to use the music each reliever has.

"Here Comes the Boom!" rocks over all others.

doublem23
09-03-2008, 10:38 PM
"Here Comes the Boom!" rocks over all others.

Saliva is poser rock.

TheOldRoman
09-03-2008, 10:39 PM
Saliva is poser rock.1.:club:
2. P.O.D. does that song.

chisoxfanatic
09-03-2008, 10:40 PM
Saliva is poser rock.
Saliva????? Isn't it POD?

doublem23
09-03-2008, 10:46 PM
1.:club:
2. P.O.D. does that song.

Saliva????? Isn't it POD?

Sorry, I sometimes mix up my crappy, late-90's-to-present faux rock bands. I was thinking of "Click, Click, Boom."

Just to be clear, POD and Saliva both suck.

chisoxfanatic
09-03-2008, 10:47 PM
Sorry, I sometimes mix up my crappy, late-90's-to-present faux rock bands. I was thinking of "Click, Click, Boom."

Just to be clear, POD and Saliva both suck.
I just got that song out of my head, doub!!! :angry:

doublem23
09-03-2008, 10:55 PM
I just got that song out of my head, doub!!! :angry:

K47W0UTq_9o

TDog
09-03-2008, 11:12 PM
Around 1970, I read an article about a relief pitcher with a better idea for evaluating the effectiveness of relievers. It might have been Tug McGraw. I don't remember the details, but it was a plus-minus system that took into account inherited runners that scored as well as a pitcher's own success. Jenks today, for example, came in with two runners on base and neither of them scored before he faced the minimum in the ninth.

The pitcher was expressing dissatisfaction with ERA, which didn't take into account inherited runners allowed, and, perhaps worse, charged relievers with earned runs for issuing intentional walks before turning over the game to other pitchers who allowed them to score. Wins and losses, of course, are arbitrary for relievers. The win vultures are rarely the best relievers on a staff. Saves, at the time, were awarded to pitchers who finished wins, regardless of the score. Even advocates of the stat found the criteria dissatisfying.

I don't think anyone today would argue that Jenks' outing was superior today -- officially a "tough save." Saying he allowed only one baserunner in 1.2 innings doesn't really describe the quality of his effort. Saying the Indians hit .250 against him today certainly doesn't

Of course, noting that Thornton came in with two inherited runners that didn't score fails to tell the story of his day. I don't believe there is one stat line that will adequately track any pitcher's performance. And, of course, stats tell you where a pitcher has been, not what a pitcher do.

Lefty34
09-03-2008, 11:39 PM
I've been thinking of this a while, actually. ERA, WHIP, record, holds, and saves are pretty standard statistics for relief pitchers. You'll see all of those numbers in the stat box whenever a relief pitcher enters the game. I, however, never have seen them list anything about how many inherited runners they allow to score, which I think is a very telling stat in determining the quality of a relief pitcher (they're paid to not only get their guys out, but to strand runners as well). You'll rarely even hear the commentators discuss it either. I'm wondering why not as much of a big deal is made out of this. It should be one of the major relief pitching stats, or at least something that deals with it.

I don't like the inherited runs scored statistic because it is still does not eliminate the defense portion of pitching. I would look more at a pitcher's Defense-Neutral ERA as well as his Defense-neutral ERA (different statistic from dERA) and Normalized Runs Allowed numbers. Though complicated, I believe they do a good job of eliminating the defense and luck that a pitcher has to deal with, rather than worrying about things a pitcher has little or no control over (like BABIP). Oh yeah, and don't forget about Left One Out GuY (LOOGY).

TheOldRoman
09-03-2008, 11:44 PM
sorry, i sometimes mix up my crappy, late-90's-to-present faux rock bands. I was thinking of "click, click, boom."

just to be clear, pod and saliva both suck.Well, I guess that settles it.:rolleyes:

Foulke You
09-03-2008, 11:59 PM
I prefer to use the music each reliever has.

"Here Comes the Boom!" rocks over all others.
While I do enjoy Bobby's current "Here Comes the Boom!" music, I did like that he briefly used "Time To Play the Game" by Motorhead when he first got called up. Now that song rocks! :cool:

Lefty34
09-04-2008, 12:17 AM
While I do enjoy Bobby's current "Here Comes the Boom!" music, I did like that he briefly used "Time To Play the Game" by Motorhead when he first got called up. Now that song rocks! :cool:

Umm I think you guys are forgetting Derrick Turnbow's use of Fuel for Fire by Metallica. Too bad he sucked.

doublem23
09-04-2008, 08:21 AM
I don't like the inherited runs scored statistic because it is still does not eliminate the defense portion of pitching. I would look more at a pitcher's Defense-Neutral ERA as well as his Defense-neutral ERA (different statistic from dERA) and Normalized Runs Allowed numbers. Though complicated, I believe they do a good job of eliminating the defense and luck that a pitcher has to deal with, rather than worrying about things a pitcher has little or no control over (like BABIP). Oh yeah, and don't forget about Left One Out GuY (LOOGY).

Defense, for the most part, I believe evens out over enough time.

That, and why would you want to remove "defense?" It's 1/3 of the game. That would be like recording rushing yards in football but trying to take out the effects of blocking. :scratch:

BringBackBlkJack
09-04-2008, 08:27 AM
I've got two points.

1. Saliva and P.O.D. do, indeed, suck.

2. I think inherited earned run average should be a "money" stat for relievers. It should be listed up there with ERA, holds and saves, and W-L shouldn't be anywhere in the mix. The only issue with IERA is how to compute it fairly. It cannot be even across the board because a reliever coming in with two outs and a guy on 1st would get the same credit as a guy coming in with no outs and a guy on 3rd. Mathematicians, get on it!

Lefty34
09-04-2008, 01:58 PM
Defense, for the most part, I believe evens out over enough time.

That, and why would you want to remove "defense?" It's 1/3 of the game. That would be like recording rushing yards in football but trying to take out the effects of blocking. :scratch:

No it really wouldn't, because blocking has a direct affect on how a back can run and what he can do. However Voros McCracken, along with many others, have gone through reams of data and shown that a pitcher has very little or no control over what a ball does when it is put into play (his BABIP) and he very usually just relies on his defense and luck. We know this is true because when you look at a pitcher's BABIP from year to year, it will most likely being extremely volatile, it will be sky high one year then at a career low the next and so on and so forth.

So why would you want to include defense, if all it is going to do is to cloud whether a pitcher is actually good at what he can control, like walks, strikeouts, HBP's and HR's. There should be no call to reward a pitcher who played in front of a superb defense, the defense should be awarded for that.

doublem23
09-04-2008, 02:28 PM
No it really wouldn't, because blocking has a direct affect on how a back can run and what he can do. However Voros McCracken, along with many others, have gone through reams of data and shown that a pitcher has very little or no control over what a ball does when it is put into play (his BABIP) and he very usually just relies on his defense and luck. We know this is true because when you look at a pitcher's BABIP from year to year, it will most likely being extremely volatile, it will be sky high one year then at a career low the next and so on and so forth.

But a RB can't control the holes that his line makes, or whether or not they properly execute the scheme for him, pretty much in the same way that a pitcher can't control what happens to a ball when it's put in play.

In any way, pitching and defense are intrinsically connected; a pitcher's success depends on his defense and vice versa, trying to remove one from the other is impossible. At best, you'll end up with a stat like fielding percentage, which seems nice but is ultimately worthless.

skottyj242
09-04-2008, 02:52 PM
Payable On Death.

Lefty34
09-04-2008, 02:58 PM
But a RB can't control the holes that his line makes, or whether or not they properly execute the scheme for him, pretty much in the same way that a pitcher can't control what happens to a ball when it's put in play.

In any way, pitching and defense are intrinsically connected; a pitcher's success depends on his defense and vice versa, trying to remove one from the other is impossible. At best, you'll end up with a stat like fielding percentage, which seems nice but is ultimately worthless.

No it is worth it to separate a pitcher from his defense because otherwise you would sign, trade for, or bring up a guy that had a good defense behind him, and not get that defense he played in front of. And I wish we would stop with the football analogies because football and baseball are two completely different sports. Football is more of an all-encompassing team sport (almost everyone involved on every play, working toward a common goal) while baseball starts with a one on one competition, and then it is usually a defense v. runner(s) competition.

Unless the ball is hit right at him, a pitcher has, once again, little or no control over what it does in the field (though knuckleballers and other "trick" pitchers seem to have a little control over what kind of ball is hit); after that, he is more often than not out of the play. So again, why should we reward relievers and pitchers in general for having a good defense?

Let's say I pitch for team A and you pitch for team B, and we both get roughly 100 IP and both have an ERA of around 3.00. Also, we both are amazing at striking people out (everyone that doesn't get a hit) and we keep our BB's to the bare minimum of 0. Now, Team A is comprised of the best defenders you can think of, and as a reliever my BABIP is .000. However, my HR rate is hight at 3 HR/ 9 IP. You, on the other hand, pitch for team B, which is comprised of fielders so slow and out of position on every pitch that your BABIP is pretty high. However you still K batters at the same rate and give up little or no home runs and walk just as few, and finish the season with a 3.00 ERA or thereabouts. Who is the better pitcher? I finished the season with a similar ERA , but only because I pitched in front of a vastly superior defense, while you K'd people at the same rate as I did, you BB/9 was the same as mine, and you gave up far fewer HR's, but your defense would make Manny Ramirez say "damn, that's bad." Who would you take?

That is why pitching statistics that do not neutralize the defense can be suspect, and why stats that reduce or minimize the defense aspect of the game can be more telling of a pitcher's true quality.

doublem23
09-04-2008, 03:08 PM
That is why pitching statistics that do not neutralize the defense can be suspect, and why stats that reduce or minimize the defense aspect of the game can be more telling of a pitcher's true quality.

I'm not saying that trying to determine a pitcher's worth independent of his defense isn't a worthwhile task, however, it's better achieved relying on more proven stats such as ERA, WHIP, K ratios, BABIP, etc. than trying to formulate some defense-free pitching statistic. We've covered, numerous times, how pathetically insufficient all defensive stats are, therefore any new stat designed to judge a pitcher minus the effects of his defense would be naturally flawed since it already based on other crappy data.

The football analogies, BTW, are good in this case because baseball defense resembles football more than any other aspect of baseball because it is a team effort in which a player's individual skills are often totally not quantifiable.

Lefty34
09-04-2008, 03:39 PM
I'm not saying that trying to determine a pitcher's worth independent of his defense isn't a worthwhile task, however, it's better achieved relying on more proven stats such as ERA, WHIP, K ratios, BABIP, etc. than trying to formulate some defense-free pitching statistic. We've covered, numerous times, how pathetically insufficient all defensive stats are, therefore any new stat designed to judge a pitcher minus the effects of his defense would be naturally flawed since it already based on other crappy data.

The football analogies, BTW, are good in this case because baseball defense resembles football more than any other aspect of baseball because it is a team effort in which a player's individual skills are often totally not quantifiable.

But it has been shown (numerous times) that a pitcher has little or no control over what happens to a ball once it is put into play and/or how his defense reacts. DIPS was created to neutralize the defense element of pitching, and IMO they do a really god job. I am not opposed to WHIP and K ratios, but ERA is a very suspect stat for the aforementioned reasons, and BABIP can be used to show whether a pitcher has been lucky or unlucky on defense (almost the same as using DERA and NRA) and whether his future appearances will tend to regress towards the mean.

doublem23
09-04-2008, 04:14 PM
But it has been shown (numerous times) that a pitcher has little or no control over what happens to a ball once it is put into play and/or how his defense reacts. DIPS was created to neutralize the defense element of pitching, and IMO they do a really god job. I am not opposed to WHIP and K ratios, but ERA is a very suspect stat for the aforementioned reasons, and BABIP can be used to show whether a pitcher has been lucky or unlucky on defense (almost the same as using DERA and NRA) and whether his future appearances will tend to regress towards the mean.

Per usual, the disagreement comes from how much stock you put into a lot of these sabermetric statistics. If you believe that they're great ways to evaluate players, you'll love DIPS; if you're wary of them, you won't. I mean, have you seen the formula fo dERA? It's ridiculous, and basically doesn't shed any new light (IMO) one could conclude by just analyzing the individual stats (HR rate, K, BB, etc.) individually.

The other thing that makes me leary of DIPS is that I believe, to some extent, that a pitcher does have some control over what happens to balls hit in play off them. Obviously defense is a much greater factor, but to suggest that it's all pure luck is an oversimplification. The ability to jam hitters, buckle their knees, or induce a ground ball in certain situations is at least partially (IMO) reflective of a pitcher's skill.

kjhanson
09-04-2008, 04:34 PM
But it has been shown (numerous times) that a pitcher has little or no control over what happens to a ball once it is put into play and/or how his defense reacts. DIPS was created to neutralize the defense element of pitching, and IMO they do a really god job. I am not opposed to WHIP and K ratios, but ERA is a very suspect stat for the aforementioned reasons, and BABIP can be used to show whether a pitcher has been lucky or unlucky on defense (almost the same as using DERA and NRA) and whether his future appearances will tend to regress towards the mean.

I vehemently disagree with this statement. BABIP is not a good stand-alone measure of a pitcher's effectiveness, but combined with SO:BB and a few other measures, it's extremely useful. However, what it doesn't quantify is luck. If a ball is hit sharply up the middle, that is called talent (for the hitter), not bad luck (for the pitcher). If a ball is hit weakly to short, that is called bad talent (for the hitter), not good luck (for the pitcher). If a ball is hit softly to right-field and the rightfielder drops it, that is called bad talent (for the hitter), not bad luck (for the pitcher) - and thus it is scored an error.

In other words, 99.9% (made up, of course) of what happens on a baseball field is based on talent, not luck. That is why the lesser-talented teams (Pirates, Nationals, Royals) are at the bottom of their divisions, and the moe talented teams (Angels, Cubs, White Sox) are at the top of their division.

Bad calls, Injuries and Situations (i.e: bad scheduling) are elements where luck may or may not be involved, but most everything that happens between the chalk is not subject to such ambiguity.

I haven't seen the data that you reference that heeds volatile BABIP numbers. However, I would be willing to bet that the most successful pitchers and the least successful pitchers had the most consistent BABIP.

kjhanson
09-04-2008, 04:46 PM
and whether his future appearances will tend to regress towards the mean.

There is no such thing as "regression towards the mean" in baseball. Why you ask? Because there is no calculable Expectation for any statistic in baseball. The "expected" mean for any statistic for one player may be increasing, decreasing, or static at any given time.

Example: You flip a coin. What are the chances it lands on heads? 50%. A young ballplayer comes to bat in his rookie season. What are the chances he gets a hit? He's hitting .270, so you're likely to say 27%. OK, the following year that player comes up to bat. What are the chances he gets a hit? He's hitting .310; so is it 31% or is it lower than 27%, because he's bound to "regress to his mean". But wait, what if he's a smarter, stronger hitter this year. Then what? His so-called mean/expectation will have changed to a higher number and may gradually increase throughout the rest of his career. It obviously has to do with independence. But in the coin flip, each independent scenario will result in a 50/50 shot. Each independent at-bat will have a 30/70 or a 40/60 or a 10/90 -- but because it's highly dependent on the changing environment, there's no way to calculate an expectation, and thus regressing to the mean is impossible!!!

Lefty34
09-04-2008, 04:55 PM
I vehemently disagree with this statement. BABIP is not a good stand-alone measure of a pitcher's effectiveness, but combined with SO:BB and a few other measures, it's extremely useful. However, what it doesn't quantify is luck. If a ball is hit sharply up the middle, that is called talent (for the hitter), not bad luck (for the pitcher). If a ball is hit weakly to short, that is called bad talent (for the hitter), not good luck (for the pitcher). If a ball is hit softly to right-field and the rightfielder drops it, that is called bad talent (for the hitter), not bad luck (for the pitcher) - and thus it is scored an error.

In other words, 99.9% (made up, of course) of what happens on a baseball field is based on talent, not luck. That is why the lesser-talented teams (Pirates, Nationals, Royals) are at the bottom of their divisions, and the moe talented teams (Angels, Cubs, White Sox) are at the top of their division.

Bad calls, Injuries and Situations (i.e: bad scheduling) are elements where luck may or may not be involved, but most everything that happens between the chalk is not subject to such ambiguity.

I haven't seen the data that you reference that heeds volatile BABIP numbers. However, I would be willing to bet that the most successful pitchers and the least successful pitchers had the most consistent BABIP.

Would Nolan Ryan (http://www.fangraphs.com/graphs.aspx?playerid=1011348&position=P&page=7&type=full) Work? How about Roger Clemens (http://www.fangraphs.com/graphs.aspx?playerid=1011348&position=P&page=7&type=full)? Not yet? How about Tom Glavine (http://www.fangraphs.com/graphs.aspx?playerid=1011348&position=P&page=7&type=full), Greg Maddux (http://www.fangraphs.com/graphs.aspx?playerid=104&position=P&page=7&type=full), and Roy Halladay (http://www.fangraphs.com/graphs.aspx?playerid=1303&position=P&page=7&type=full)? Do those work for you? Does that illustrate the luck part of BABIP for you? Or are those all "skill" graphs?

doublem23
09-04-2008, 04:58 PM
Would Nolan Ryan (http://www.fangraphs.com/graphs.aspx?playerid=1011348&position=P&page=7&type=full) Work? How about Roger Clemens (http://www.fangraphs.com/graphs.aspx?playerid=1011348&position=P&page=7&type=full)? Not yet? How about Tom Glavine (http://www.fangraphs.com/graphs.aspx?playerid=1011348&position=P&page=7&type=full), Greg Maddux (http://www.fangraphs.com/graphs.aspx?playerid=104&position=P&page=7&type=full), and Roy Halladay (http://www.fangraphs.com/graphs.aspx?playerid=1303&position=P&page=7&type=full)? Do those work for you? Does that illustrate the luck part of BABIP for you? Or are those all "skill" graphs?

:scratch:

Those are all pretty consisent.

I was expecting wild, crazy swings... What's the greatest standard deviation any of these graphs show?

Lefty34
09-04-2008, 05:07 PM
There is no such thing as "regression towards the mean" in baseball. Why you ask? Because there is no calculable Expectation for any statistic in baseball. The "expected" mean for any statistic for one player may be increasing, decreasing, or static at any given time.

Example: You flip a coin. What are the chances it lands on heads? 50%. A young ballplayer comes to bat in his rookie season. What are the chances he gets a hit? He's hitting .270, so you're likely to say 27%. OK, the following year that player comes up to bat. What are the chances he gets a hit? He's hitting .310; so is it 31% or is it lower than 27%, because he's bound to "regress to his mean". But wait, what if he's a smarter, stronger hitter this year. Then what? His so-called mean/expectation will have changed to a higher number and may gradually increase throughout the rest of his career. It obviously has to do with independence. But in the coin flip, each independent scenario will result in a 50/50 shot. Each independent at-bat will have a 30/70 or a 40/60 or a 10/90 -- but because it's highly dependent on the changing environment, there's no way to calculate an expectation, and thus regressing to the mean is impossible!!!

Not when you use it for a statistic like BABIP, because the pool you are drawing from is all pitchers in the league. A typical BABIP for a pitcher is about .290, which means that a pitcher with a BABIP higher or lower can expect his numbers in the future to regress towards the mean. So let's say your prototypical rookie is hitting .340 in the majors (probably a bit of an outlier for rookies) he can expect his next season or group of AB's to have a mean of the league average. Or to use a different example, let's say 100 kids take a 100 question true-false test, with a mean of 50 in the results. Some will score a lot higher and some will score a lot lower, obviously. So now let's take the lowest 20 scores and give those kids another guessing test, we can expect the results of that exam to have a mean of 50 as well, meaning they would all have regressed to the mean.

Lefty34
09-04-2008, 05:11 PM
:scratch:

Those are all pretty consisent.

I was expecting wild, crazy swings... What's the greatest standard deviation any of these graphs show?

They don't give the SD's of the graph, and I'm really too lazy to calculate it for a bunch of people on a forum who will disagree with what I say anyways.

But doesn't that seem pretty inconsistent for something that is supposed to be a "skill"? Look at the actual numbers as they go down the column and pretend they were something like Batting Average. Wouldn't that lead you to believe that the hitter producing those stats was inconsistent at best? I mean come on.

doublem23
09-04-2008, 11:11 PM
But doesn't that seem pretty inconsistent for something that is supposed to be a "skill"? Look at the actual numbers as they go down the column and pretend they were something like Batting Average. Wouldn't that lead you to believe that the hitter producing those stats was inconsistent at best? I mean come on.

You're the one who just this week posted the quote from Bull Durham about a 50-point season swing being just 1 extra hit per week? No one is discounting that there is some luck involved, but by large, good or poor baseball is the product of skill. It's not just a game of random events. I guess we just read the graphs differently, I see:

a) Though the yearly BABIPs tend to sway up and down more violently, they pretty much follow the MLB average over the course of their careers.

b) Again, although there is yearly discrepancy, these stats never go into the "bad' category. These "wild swings" that your claiming aren't really that wild... Talk to me when you find a solid pitcher whose BABIP jumps around by more than what looks like 50 BA points or so.

c) Compared to some great hitters, here's what their yearly BA graphs look like: Mickey Mantle (http://www.fangraphs.com/graphs.aspx?playerid=1008082&position=OF&page=0&type=full), Ted Williams (http://www.fangraphs.com/graphs.aspx?playerid=1014040&position=OF&page=0&type=full), Tony Gwynn (http://www.fangraphs.com/graphs.aspx?playerid=1005166&position=OF&page=0&type=full), Rickey Henderson (http://www.fangraphs.com/graphs.aspx?playerid=194&position=OF&page=0&type=full), and Derek Jeter (http://www.fangraphs.com/graphs.aspx?playerid=826&position=SS&page=0&type=full). These BA swings are just as volitale as your BABIP graphs... Is hitting all pure luck, too?

kjhanson
09-05-2008, 09:57 AM
Not when you use it for a statistic like BABIP, because the pool you are drawing from is all pitchers in the league. A typical BABIP for a pitcher is about .290, which means that a pitcher with a BABIP higher or lower can expect his numbers in the future to regress towards the mean. So let's say your prototypical rookie is hitting .340 in the majors (probably a bit of an outlier for rookies) he can expect his next season or group of AB's to have a mean of the league average. Or to use a different example, let's say 100 kids take a 100 question true-false test, with a mean of 50 in the results. Some will score a lot higher and some will score a lot lower, obviously. So now let's take the lowest 20 scores and give those kids another guessing test, we can expect the results of that exam to have a mean of 50 as well, meaning they would all have regressed to the mean.

Albert Pujols hit .329 in his rookie year. Why didn't his average "regress" the next year? Did he get lucky, for seemingly the second straight year? No. He's more talented than everyone else, which means he behaves differently. And because he behaves differently, you can't explain his performance relative to some expectation. At the end of 2001 you could have assumed that his average would drop, but it would be impossible for you to quantify and then prove that any statisitcal drop or "regression" was imminent.

As far as your children taking the test example.... If 100 kids take a true/false test with a mean of 50, that's completely fine. And then you say some will score higher and some will score lower. That's fine as well. And then you give the test to the 20 lowest scorers and you expect their mean to be 50 because it is a guessing test. Was it a guessing test for your original sample? You say it was true/false, which would indicate that some type of skill/knowledge would lead to a right answer. In which case, the 20 lowest scorers, would probably also score low on a similar test. However, if it was a guessing game the first time around, I would expect those low scorers to regress towards the expectation of 50 the second time around. Why? Because if it's truly a guessing game, there is a calculable expectation of 50 as a score. How does this apply to baseball though? Do you truly believe that every time someone hits a baseball with a bat neither the hitter nor the pitcher have any control over what happens to it? If you honestly believe that, I will buy your box of Kleenex when you finally come to the realization that it's not true!

How come Barry Zito's BABIP was .248 in 2002 when he won 22 games? And why is it .301 this year? It's not all luck, my friend.

Lefty34
09-05-2008, 10:34 AM
Albert Pujols hit .329 in his rookie year. Why didn't his average "regress" the next year? Did he get lucky, for seemingly the second straight year? No. He's more talented than everyone else, which means he behaves differently. And because he behaves differently, you can't explain his performance relative to some expectation. At the end of 2001 you could have assumed that his average would drop, but it would be impossible for you to quantify and then prove that any statisitcal drop or "regression" was imminent.

As far as your children taking the test example.... If 100 kids take a true/false test with a mean of 50, that's completely fine. And then you say some will score higher and some will score lower. That's fine as well. And then you give the test to the 20 lowest scorers and you expect their mean to be 50 because it is a guessing test. Was it a guessing test for your original sample? You say it was true/false, which would indicate that some type of skill/knowledge would lead to a right answer. In which case, the 20 lowest scorers, would probably also score low on a similar test. However, if it was a guessing game the first time around, I would expect those low scorers to regress towards the expectation of 50 the second time around. Why? Because if it's truly a guessing game, there is a calculable expectation of 50 as a score. How does this apply to baseball though?

How come Barry Zito's BABIP was .248 in 2002 when he won 22 games? And why is it .301 this year? It's not all luck, my friend.

Yes it was a guessing test all throughout the example, and in my posting frenzy I must have failed to mention that at the beginning.

And my BA example was in bad taste, because of course there is tremendous skill involved in hitting, and a hitter truly does have an influence on what his batted balls do and where they go. However the pitcher the graphs should clearly illustrate that a pitcher has little or no control over his BABIP.

Do you truly believe that every time someone hits a baseball with a bat neither the hitter nor the pitcher have any control over what happens to it? If you honestly believe that, I will buy your box of Kleenex when you finally come to the realization that it's not true!

Did you really just ask me this? Moreover, do you actually believe in your position so much that you are willing to abandon reason? Granted certain trick pitchers have the ability to induce certain types of batted balls, but once the balls are put into play, they have little or no control over the outcome of the play. Are you really disputing this? I mean, other than balls batted right at the pitcher and the like, are you really saying that the pitcher has some control over how his fielders react and how they put a play on his batted pitches? Take a step back, remove a certain object from rectal cavity, and look at what you are disputing.

And get your own damn Kleenex box.

Lefty34
09-05-2008, 10:47 AM
And in 2004 when Roy Oswalt went 20-10, his BABIP was .320. So maybe there is a lot more luck involved than you want to believe...my friend.

kjhanson
09-05-2008, 11:15 AM
Did you really just ask me this? Moreover, do you actually believe in your position so much that you are willing to abandon reason? Granted certain trick pitchers have the ability to induce certain types of batted balls, but once the balls are put into play, they have little or no control over the outcome of the play. Are you really disputing this? I mean, other than balls batted right at the pitcher and the like, are you really saying that the pitcher has some control over how his fielders react and how they put a play on his batted pitches?

I never said that the pitcher can control how his fielders field, where they are positioned, how much range they have, and their overall defensive ability. That's absurd. But what it isn't, is luck.

What I'm saying is that a pitcher has control great control over whether the ball is hit in the air, on the ground, or on a line. And when the ball is hit on a line, it is more likely to be a base-hit. In your Roy Oswalt season, 22.2% of hits off of him were line-drives - one of the highest figures in his career. Percentage of line drives is highly correlated with BABIP. And I would argue that pitchers who give up fewer line-drives are better than those who give up more line-drives.

In my original argument I noted that BABIP is not a good stand-alone measure of a pitcher. Do you want to know what measure does neutralize defense somewhat? SLGBIP (you have to include home runs). Though BABIP is confounded in SLGBIP, the addition of adding total bases to calculation holds defense somewhat constant. A home run is a home run, whether the outfielder falls down, is one-legged, etc. (except the rare home-run saving catch), and a double is usually a double.

Lefty34
09-05-2008, 11:37 AM
I never said that the pitcher can control how his fielders field, where they are positioned, how much range they have, and their overall defensive ability. That's absurd. But what it isn't, is luck.

What I'm saying is that a pitcher has control great control over whether the ball is hit in the air, on the ground, or on a line. And when the ball is hit on a line, it is more likely to be a base-hit. In your Roy Oswalt season, 22.2% of hits off of him were line-drives - one of the highest figures in his career. Percentage of line drives is highly correlated with BABIP. And I would argue that pitchers who give up fewer line-drives are better than those who give up more line-drives.

In my original argument I noted that BABIP is not a good stand-alone measure of a pitcher. Do you want to know what measure does neutralize defense somewhat? SLGBIP (you have to include home runs). Though BABIP is confounded in SLGBIP, the addition of adding total bases to calculation holds defense somewhat constant. A home run is a home run, whether the outfielder falls down, is one-legged, etc. (except the rare home-run saving catch), and a double is usually a double.

Yeah I have often liked the SLG factor as a measure of pitchers, but for SLGBIP, why are HR's included? They are never technically in-play, so can't we just rule them out and then adjust the range for good v. bad SLGBIP?

chaerulez
09-05-2008, 03:18 PM
I don't know what you guys are arguing about really, it seems everyone agrees that a pitcher has no control over balls in play. Therefore truly trying to evaluate a pitcher you have to focus on the stats he has at least some control over such as BB, K, HR, and GB/FB ratio.

kjhanson
09-05-2008, 04:07 PM
Yeah I have often liked the SLG factor as a measure of pitchers, but for SLGBIP, why are HR's included? They are never technically in-play, so can't we just rule them out and then adjust the range for good v. bad SLGBIP?

The reason you have to include HRs into SLGBIP is because many pitchers give them up at a very disproportional rate. So if you exclude them, you're punishing guys who keep the ball in the park, and aiding those who let 'em fly. Here's a good example of the difference:

Matt Morris, 2004 : 15-10, 2.34 K:BB
SLGBIP (HRs not included): .351

Brandon Webb, 2007: 18-10, 2.69 K:BB
SLGBIP (HRs not included): .365

Both of these guys had seemingly similar seasons, yet 99 out of 100 people would take the 2007 Brandon Webb (3.01 ERA) over the 2004 Matt Morris (4.72 ERA).

Here are their SLGBIP with home runs included:
MM: .550
BW: .428

Lefty34
09-06-2008, 02:44 AM
The reason you have to include HRs into SLGBIP is because many pitchers give them up at a very disproportional rate. So if you exclude them, you're punishing guys who keep the ball in the park, and aiding those who let 'em fly. Here's a good example of the difference:

Matt Morris, 2004 : 15-10, 2.34 K:BB
SLGBIP (HRs not included): .351

Brandon Webb, 2007: 18-10, 2.69 K:BB
SLGBIP (HRs not included): .365

Both of these guys had seemingly similar seasons, yet 99 out of 100 people would take the 2007 Brandon Webb (3.01 ERA) over the 2004 Matt Morris (4.72 ERA).

Here are their SLGBIP with home runs included:
MM: .550
BW: .428

Ok I got ya, I was just curious because HR's are not included in BABIP and the other Balls In Play stats because they are not considered "in play". But that makes sense to me.

Mohoney
09-10-2008, 09:32 PM
Maybe MLB can tweak runs allowed to account for quarter runs. If a pitcher leaves a game with a runner on 1st, and the runner scores while a different pitcher is pitching, the first pitcher gets charged with a quarter of a run and the second pitcher gets charged with three quarters of a run. For more than 2 pitchers, every base that a runner advances is a quarter of a run for the pitcher that was in the game when he advanced that particular base.

Lefty34
09-10-2008, 10:01 PM
Maybe MLB can tweak runs allowed to account for quarter runs. If a pitcher leaves a game with a runner on 1st, and the runner scores while a different pitcher is pitching, the first pitcher gets charged with a quarter of a run and the second pitcher gets charged with three quarters of a run. For more than 2 pitchers, every base that a runner advances is a quarter of a run for the pitcher that was in the game when he advanced that particular base.

Well walks still aren't counted as official AB's so...good luck.