PDA

View Full Version : PECOTA Projections Pick White Sox 3rd in AL Central


SoxFan48
03-22-2005, 09:51 AM
AL Central

W L PCT
Twins 86 76 .533
Indians 85 77 .525
White Sox 80 82 .497
Tigers 76 86 .467
Royals 69 93 .427

http://www.baseballprospectus.com/article.php?articleid=3836

jabrch
03-22-2005, 10:05 AM
sub-.500? I just don't see it. Sorry.

Ol' No. 2
03-22-2005, 10:35 AM
Good thing PECOTA projections aren't any better than just picking last year's winners to repeat.

veeter
03-22-2005, 10:51 AM
I think we should all be used to this stuff. Since these guys are always wrong, this is a good sign.

ma_deuce
03-22-2005, 10:52 AM
:cleo

"I predict that PECOTA overemphasises the effect of US Cellular being a "hitter's park," mon. No worries."

SoxFan48
03-22-2005, 10:53 AM
However, PECOTA does have a track record of capturing mean (50% probabilty) performances of players. Of course, when players move to 80-90% or down to 10-20%, then different outcomes occur.

Read the attached article on see how PECOTA translates into projected wins--84 wins last year, down to 78 as Mags departs, back to 82 with the signing of Dye, down to 77 with the infamous Lee-Podsednik trade, ticks up 78 with the signing of Hernandez, up to 80 with the signing of Pierzyntski, between 80-81 with the signing of Iguichi.

It is a methodology which people are free to accept or reject. But it is a methodology based on proven outcomes and is better starting point for discussion than "I just don't buy that".

Ol' No. 2
03-22-2005, 10:55 AM
However, PECOTA does have a track record of capturing mean (50% probabilty) performances of players. Of course, when players move to 80-90% or down to 10-20%, then different outcomes occur.

Read the attached article on see how PECOTA translates into projected wins--84 wins last year, down to 78 as Mags departs, back to 82 with the signing of Dye, down to 77 with the infamous Lee-Podsednik trade, ticks up 78 with the signing of Hernandez, up to 80 with the signing of Pierzyntski, between 80-81 with the signing of Iguichi.

It is a methodology which people are free to accept or reject. But it is a methodology based on proven outcomes and is better starting point for discussion than "I just don't buy that".Except the error bands on all those projections is large enough that the end result is that they're no better than just picking who won last year to repeat.

SoxFan48
03-22-2005, 10:56 AM
I think we should all be used to this stuff. Since these guys are always wrong, this is a good sign.

And this board is always right? How many years in a row as the offseason consensus been that this year is the White Sox year? I would argue that based on results over the past five years, the stat boys have been more right about the White Sox than the hopeful optimists.

ma_deuce
03-22-2005, 11:00 AM
AL Central

W L PCT
Indians 85 77 .525

:roflmao:

This alone should have alerted you that PETOCA is full of it. The Jndjans suck and will continue to suck throughout 2005.

Deuce

kravdog
03-22-2005, 11:03 AM
[QUOTE=SoxFan48]
Read the attached article on see how PECOTA translates into projected wins--84 wins last year, down to 78 as Mags departs, QUOTE]

So, according to PECOTA, Maggs was worth 6 wins last season in just 52 games w/ 202 ab's? With that logic, uncle jerry should have resigned Maggs for $250 mil

Lip Man 1
03-22-2005, 11:21 AM
Correct me if I'm wrong but isn't this the site that explained how it was impossible for the Twins to have won the division the past three years because of things like runs scored versus runs allowed and other factors?

Basically they inferred that the Twins winning was a fluke because the statistical numbers couldn't possibly add up to victory.

So much for the 'reliability' of this site eh?

Bottom line...baseball is a game played by athletes, not eggheads.... when a statistical formula is designed to accurately measure heart, guts, ability and the injury factor then I'll pay it some heed.

Lip

DrCrawdad
03-22-2005, 11:30 AM
Correct me if I'm wrong but isn't this the site that explained how it was impossible for the Twins to have won the division the past three years because of things like runs scored versus runs allowed and other factors?

Basically they inferred that the Twins winning was a fluke because the statistical numbers couldn't possibly add up to victory.

So much for the 'reliability' of this site eh?

Bottom line...baseball is a game played by athletes, not eggheads.... when a statistical formula is designed to accurately measure heart, guts, ability and the injury factor then I'll pay it some heed.

Lip

I took out a copy of Canseco's book from the local library. Canseco had a couple somewhat funny observations about stats guys. In essence Canseco goes along with your bottom line.

santo=dorf
03-22-2005, 11:34 AM
Lip, you are thinking of the pythagorean (http://www.baseball-reference.com/about/faq.shtml#pyth) W-L.

Pythagorean winning percentage is an estimate of a team's winning percentage given their runs scored and runs allowed. Developed by Bill James, it can tell you when teams were a bit lucky or unlucky. It is calculated by


(Runs Scored)^1.83/[(Runs Scored)^1.83 + (Runs Allowed)^1.83]

The traditional formula uses an exponent of two, but this has proven to be a little more accurate.


:rolleyes: :whatever:

pythagorean records:
White Sox:
2002: 86-76
2003: 88-74
2004: 84-78

Twins:
2002: 86-75
2003: 85-77
2004: 87-75

Lip Man 1
03-22-2005, 11:44 AM
Dorf:

Thank you. You are correct. Another fine site for accurate predictions!

Lip

SoxFan48
03-22-2005, 04:58 PM
Bottom line...baseball is a game played by athletes, not eggheads.... when a statistical formula is designed to accurately measure heart, guts, ability and the injury factor then I'll pay it some heed.

Lip

Yup, you know a person has run out of thoughts when the words heart and guts appear.

NSSoxFan
03-22-2005, 05:16 PM
I don't like math.

I don't like it, I can't stand it and it does not belong in baseball.

Dan H
03-22-2005, 05:19 PM
I will never understand why some fans get upset when some expert doesn't pick the Sox to win. Predictions are irrevelant. Results are what counts. And face it. The results for the Sox during the past few years have not been good. Unless you like win totals in the low to mid 80's.

Daver
03-22-2005, 05:27 PM
Frank Zappa picked the Sox to finish fourth, he is at least as reliable as PECOTA.

veeter
03-22-2005, 05:57 PM
And this board is always right? How many years in a row as the offseason consensus been that this year is the White Sox year? I would argue that based on results over the past five years, the stat boys have been more right about the White Sox than the hopeful optimists. I didn't say this board is always right did I? The thing I can't stand about the predictions of these "experts" is that their reasons are always way off. All of them say the Sox have no power now. You and I know that's completely false. These predictors are outsiders that maybe see 5 games a year of the Sox but watch the highlights. All of us die hard Sox fans watch darn near EVERY game. That is the only way to really know the team. To say the Sox will finish third I can accept IF the reasons are valid which mostly they aren't.

Ol' No. 2
03-22-2005, 05:58 PM
I didn't say this board is always right did I? The thing I can't stand about the predictions of these "experts" is that their reasons are always way off. All of them say the Sox have no power now. You and I know that's completely false. These predictors are outsiders that maybe see 5 games a year of the Sox but watch the highlights. All of us die hard Sox fans watch darn near EVERY game. That is the only way to really know the team. To say the Sox will finish third I can accept IF the reasons are valid which mostly they aren't.Actually, the predicters in this case probably don't see 5 games a year total. But they all have calculators and absolute faith in what their calculators say.

fquaye149
03-22-2005, 06:36 PM
Frank Zappa picked the Sox to finish fourth, he is at least as reliable as PECOTA.

At least Frank knows for certain not to eat the yellow snow. Pecota would spend years crunching the numbers on the tainted snow only to come to the conclusion that we don't know for sure whether clutch hitters exist.

johnny_mostil
03-22-2005, 06:46 PM
Frank Zappa picked the Sox to finish fourth, he is at least as reliable as PECOTA.

Impressive, since he's dead.

Banix12
03-22-2005, 06:46 PM
Predictions of this nature are usually based on Raw numbers which are usually averages. Not really an accurate depiction of the highs and lows a team can go through during a season. They also can't accurately predict freak incidents that can happen which affect a teams overall performance, like Magglio getting injured last season or Rowand having a breakout season. Of course no method of predicting, even just plain guessing, can really take that into account.

Their numbers as far as predicting the sox in the past have been off because they haven't been able to depict the true nature of the white sox. Score 16 runs one night, 0 the next night with very little in between. It's difficult to make statistical predictions on a team with a tendency toward extremes. It screws up the averages and screws up the predictions.

I'm actually surprised we weren't ranked lower by them, since the Sox have so many little lingering questions going into the season that can't be statistically gauged all that well. How will Iguchi adapt to the US parks? Jermaine Dye moves into a hitters park, how does he fare? Were Rowand, Uribe and Shingo flukes? Will Konerko revert to 03? Can Joe Crede get it together? Can El Duque stay healthy? Can Contreras find consistancy? What if McCarthy comes up, what kind of monkey wrench does that throw into predictions?

If all goes well the sox can be a 100 win team. If all goes wrong it could be 100 losses. They have to go out and play the games and nothing PECOTA can say will change the outcome of actual games.

MeanFish
03-22-2005, 06:51 PM
Predictions of this nature are usually based on Raw numbers which are usually averages. Not really an accurate depiction of the highs and lows a team can go through during a season. They also can't accurately predict freak incidents that can happen which affect a teams overall performance, like Magglio getting injured last season or Rowand having a breakout season. Of course no method of predicting, even just plain guessing, can really take that into account.

Their numbers as far as predicting the sox in the past have been off because they haven't been able to depict the true nature of the white sox. Score 16 runs one night, 0 the next night with very little in between. It's difficult to make statistical predictions on a team with a tendency toward extremes. It screws up the averages and screws up the predictions.

I'm actually surprised we weren't ranked lower by them, since the Sox have so many little lingering questions going into the season that can't be statistically gauged all that well. How will Iguchi adapt to the US parks? Jermaine Dye moves into a hitters park, how does he fare? Were Rowand, Uribe and Shingo flukes? Will Konerko revert to 03? Can Joe Crede get it together? Can El Duque stay healthy? Can Contreras find consistancy? What if McCarthy comes up, what kind of monkey wrench does that throw into predictions?

If all goes well the sox can be a 100 win team. If all goes wrong it could be 100 losses. They have to go out and play the games and nothing PECOTA can say will change the outcome of actual games.

Word.

johnny_mostil
03-22-2005, 06:51 PM
However, PECOTA does have a track record of capturing mean (50% probabilty) performances of players. Of course, when players move to 80-90% or down to 10-20%, then different outcomes occur.


But PECOTA deals with injuries by spreading them around evenly, which makes it an accurate collective predictor.

Also, PECOTA doesn't know about several other factors, including platooning. Take the projection for Twin Mike Cuddyer. PECOTA doesn't know he was carefully protected from good RHPs last year. If his playing time expands his performance will suffer.

I have, on my bulletin board, last year's PECOTAs for the White Sox.

Here are two:

Takatsu_Shingo 10.44 ERA 25ip
Diaz_Felix 4.84ERA 110ip

All I can say is BRILLIANT! :gulp:

Daver
03-22-2005, 06:57 PM
Impressive, since he's dead.

About as impressive as anything PECOTA has ever done.

JRIG
03-22-2005, 07:41 PM
Actually, the predicters in this case probably don't see 5 games a year total. But they all have calculators and absolute faith in what their calculators say.

But the people who talk about heart and guts can have absolute faith in measuring those "intangible" skills?

Please. Like Hawk today saying he doesn't care about Joe Crede's .239 average because he's got the guts to get some big hits when it counts. If he sucks the other 97% of the time, it is a big deal.

They're just using stats to make a prediction. Nobody, nobody, is saying it's the be all end all of predictions. Unless, of course, it said the Sox would win it all. Then we'd love them.

JRIG
03-22-2005, 07:46 PM
I don't like math.

I don't like it, I can't stand it and it does not belong in baseball.

Then why are you quoting Frank's career batting average in your signature? You do know that's a mathematical equation of hits divided by at bats, right?

Banix12
03-22-2005, 07:54 PM
This, what we are doing on this board, is the only purpose of making preseason predictions. They are only good for arguements and prove nothing.

And in this case it's basically free advertising for the site that makes the predictions. Hopefully increasing their readership, whether you agree with them or not.

I put no stock in anybody's predictions unless they have a time machine. In which case, I want the time machine and I'll kill them to get it. And good luck catching me for murder cause I'll have a time machine... suckers!

I gotta stop watching Back to the Future 2 before I go to bed.

Dadawg_77
03-22-2005, 08:24 PM
:roflmao:

This alone should have alerted you that PETOCA is full of it. The Jndjans suck and will continue to suck throughout 2005.

Deuce

5th best offense last year that finished 80-82, with crappy pitching. This year the pitching has been improved, doesn't equal being dismissed as sucking.

Daver
03-22-2005, 08:30 PM
5th best offense last year that finished 80-82, with crappy pitching. This year the pitching has been improved, doesn't equal being dismissed as sucking.

Spoken like a true propellerhead.

Dadawg_77
03-22-2005, 08:35 PM
[QUOTE=SoxFan48]
Read the attached article on see how PECOTA translates into projected wins--84 wins last year, down to 78 as Mags departs, QUOTE]

So, according to PECOTA, Maggs was worth 6 wins last season in just 52 games w/ 202 ab's? With that logic, uncle jerry should have resigned Maggs for $250 mil

No he is worth an est. 6 wins this year.

Dadawg_77
03-22-2005, 08:37 PM
Spoken like a true propellerhead.

LOL, yeah guess we'll see how it turns out on the field. But if Tribe does better then you expect, I bet we won't he you say you were wrong about them.

Dadawg_77
03-22-2005, 08:46 PM
I didn't say this board is always right did I? The thing I can't stand about the predictions of these "experts" is that their reasons are always way off. All of them say the Sox have no power now. You and I know that's completely false. These predictors are outsiders that maybe see 5 games a year of the Sox but watch the highlights. All of us die hard Sox fans watch darn near EVERY game. That is the only way to really know the team. To say the Sox will finish third I can accept IF the reasons are valid which mostly they aren't.

Ever think you are too close to the team that you can't see its flaws. Thus any point any makes that doesn't agree with your, you assume is wrong and thus you do not take it into account, when forming your opinion?

PECOTA used the numbers which has seen everything a player did on the field did last year and can recall everything. That is more then what any spectator can do. Scouts and others project based on tools and other abilities they believe a player has. PECOTA attempts this also by its similarity scores based on a players production and body type. Since PECOTA hasn't been peer reviewed and methodlogy isn't a matter of public record, its hard to say how good of a system it actually is. However in the past couple of years, it has a decent track record, while failures which have been pointed out, it actually like Willy Mo Pena to have breakout seasons in past couple of years.

jabrch
03-22-2005, 08:49 PM
I don't like math.

I don't like it, I can't stand it and it does not belong in baseball.


Add up all the runs after nine innings and tell me who won...that's enough math for me.

Dadawg_77
03-22-2005, 08:51 PM
But PECOTA deals with injuries by spreading them around evenly, which makes it an accurate collective predictor.

Also, PECOTA doesn't know about several other factors, including platooning. Take the projection for Twin Mike Cuddyer. PECOTA doesn't know he was carefully protected from good RHPs last year. If his playing time expands his performance will suffer.

I have, on my bulletin board, last year's PECOTAs for the White Sox.

Here are two:

Takatsu_Shingo 10.44 ERA 25ip
Diaz_Felix 4.84ERA 110ip

All I can say is BRILLIANT! :gulp:

Two things, Shingo predictions were based on Japanese league production which adds to their margin of error. Also PECOTA likes pitchers who strike out more batters which Takatsu did not do in Japan. The change of the leagues made Shingo's pitches more effective since hitters have not and did not see another pitcher similar to him. That is some thing, you can not expect a computer program to pick up on.

Daver
03-22-2005, 08:59 PM
Two things, Shingo predictions were based on Japanese league production which adds to their margin of error. Also PECOTA likes pitchers who strike out more batters which Takatsu did not do in Japan. The change of the leagues made Shingo's pitches more effective since hitters have not and did not see another pitcher similar to him. That is some thing, you can not expect a computer program to pick up on.

Keep on going, I'm sure if you try you can put a propellerhead spin on every post in this thread.

Dadawg_77
03-22-2005, 09:02 PM
Add up all the runs after nine innings and tell me who won...that's enough math for me.

Baseball since its start has been about numbers. The history of the game is based on numbers, Home Run King, Hall of Fame, Most Hits in a Season. The strategy has always been about probability, sacrificing is done in the belief a runner has a higher chance to score in the new postion. Today the numbers are more complex and measure more. We question the strategy of sacrificing because it doesn't increase the probability of scoring runs, something Earl Weaver pick up on. In the era of nine figure salaries and free agency, teams need more and better ways to evaluate players to avoid breaking fans hearts and wasting millions. Sabermetrics is taking what Henry Chadwick did when he created the box score, Branch Rickey when he started to evaluate players by OBP and moving it into new era with better and more information.

Dadawg_77
03-22-2005, 09:04 PM
Keep on going, I'm sure if you try you can put a propellerhead spin on every post in this thread.

and I am sure you can say how their catcher's defense isn't that good so they won't win. I am just trying to explain things to people since I don't think they have read about how the tool does what it does. Sorry that I try to help people understand something instead of making cheap comments.

ma-gaga
03-22-2005, 09:05 PM
Keep on going, I'm sure if you try you can put a propellerhead spin on every post in this thread.

He could, but there's like 30 non-propeller heads' to every one on this board.

Eventually the overwhelming odds catch up. :cool: If only PECOTA was as accurate.

soxwon
03-22-2005, 09:07 PM
[QUOTE=SoxFan48]
Read the attached article on see how PECOTA translates into projected wins--84 wins last year, down to 78 as Mags departs, QUOTE]

So, according to PECOTA, Maggs was worth 6 wins last season in just 52 games w/ 202 ab's? With that logic, uncle jerry should have resigned Maggs for $250 mil

How is Maggs doing in ST?
Is he playing the field?

CWSGuy406
03-22-2005, 09:27 PM
I don't like math.

I don't like it, I can't stand it and it does not belong in baseball.

Heck, I'm sure glad that the Sox' way of not really utilizing some of these new numbers coming out have led to so many playoff appearances.

EDIT: And Lip, why were you bashing the Pythagorean rankings? Last year they/it were/was off by one game, 2003 by two games, and 2002 it was off by five games...

Banix12
03-22-2005, 09:42 PM
Baseball since its start has been about numbers. The history of the game is based on numbers, Home Run King, Hall of Fame, Most Hits in a Season. The strategy has always been about probability, sacrificing is done in the belief a runner has a higher chance to score in the new postion. Today the numbers are more complex and measure more. We question the strategy of sacrificing because it doesn't increase the probability of scoring runs, something Earl Weaver pick up on. In the era of nine figure salaries and free agency, teams need more and better ways to evaluate players to avoid breaking fans hearts and wasting millions. Sabermetrics is taking what Henry Chadwick did when he created the box score, Branch Rickey when he started to evaluate players by OBP and moving it into new era with better and more information.

I'm all for expanding the box score and using new tools to evaluate players. most of it is useful information, especially when evaluating individual players. Statistics shouldn't be the only tool, but it is an incredibly useful tool

However using this info to try and accurately predict the outcome of a team over an entire season is a waste of time. Too many intangibles that can't be taken into account.

Also, I don't want anyone to find a way to accurately project the season before it's played, that would make watching baseball complete waste of time. The mystery of what will happen is part of the mystique.

jeremyb1
03-22-2005, 09:46 PM
A few things...

I assume most people understand on some level that calling people names like "propeller head" and "egg head" are simply that, namecalling, and therefore have no bearing on any serious, reasoned discussion. Furthermore, DaDawg makes an excellent point that RBIs, batting average, stolen bases, and numerous other statistics have been and continue to be heavily relied on by baseball traditionalists. Even wins are a statistic and certainly everyone agrees, a critical figure in a season. With Beane, Epstein, Riccardi, and Depodesta all entrusted with budgets totaling millions, it is an unimaginable stretch to say that statistical analysis has no place in baseball. The fact that the A's are on of the game's more successful teams over the past several years and the Red Sox won the World Series last season only further emphasizes this fact.

The PECOTA projections are just that, projections. If you look up the words "projection" and "prediction" in the Webster's you'll find the terms don't indicate precise accuracy because they are examining unknowns. If virtually every baseball writer (as well as most fans on this message board) predict where teams will finish in the fall, why is it such a crime for BP to use PECOTA to attempt to do the same? If you disagree that's fine. It is difficult if not impossible for the average person to base predictions on the same amount of knowledge as PECOTA since we're talking about a gigantic database of information on baseball players throughout history, but certainly different analytical methods are not without value.

To the extent that there is misunderstanding with the projections it seems to center around the fact that what PECOTA is doing is projecting the most likely outcome based on the numbers throughout history (I am correct in assuming that there's absolutely no support behind the assertions last seasons individual performances/win totals are equally successful in projecting the current season, right?). Why people tend to assume that the PECOTA projections are meant to suggest it is extremely unlikely or impossible for the White Sox to win 81 games let alone 88 just because the projected win total is 80 is completely beyond me. When Peter Gammons picks the Twins to win the AL Central I'm sure he would not tell you he is absolutely positive in that outcome just as most of you aren't completely positive if you pick the Braves to win the NL East. It is an educated guess.

Several posters in this thread seem be attempting to argue against the PECOTA projections by stating that you have to play the games and that certain events cannot be forseen. That reality has certainly not been lost on anyone over at BP. The fact that PECOTA generates the potential for players to "breakout" and "collapse" is a means of further examining players' ability to have seasons that seemingly come out of nowhere. Furthermore the authors realize that sometimes events happen that absolutely no one plans for or sees coming, that's an obvious fact of life which is not limited to baseball. What I cannot understand is why some consider the fact that PECOTA is far from capable of perfectly predicting the future a reason to reject the system entirely and refuse to give the projections any consideration whatsoever. If I'm filling out a lottery ticket and you can tell me the first number, I'm not about to refuse to use that number just because you can't tell me the rest...

Daver
03-22-2005, 10:16 PM
A few things...

I assume most people understand on some level that calling people names like "propeller head" and "egg head" are simply that, namecalling, and therefore have no bearing on any serious, reasoned discussion. Furthermore, DaDawg makes an excellent point that RBIs, batting average, stolen bases, and numerous other statistics have been and continue to be heavily relied on by baseball traditionalists. Even wins are a statistic and certainly everyone agrees, a critical figure in a season. With Beane, Epstein, Riccardi, and Depodesta all entrusted with budgets totaling millions, it is an unimaginable stretch to say that statistical analysis has no place in baseball. The fact that the A's are on of the game's more successful teams over the past several years and the Red Sox won the World Series last season only further emphasizes this fact.

The PECOTA projections are just that, projections. If you look up the words "projection" and "prediction" in the Webster's you'll find the terms don't indicate precise accuracy because they are examining unknowns. If virtually every baseball writer (as well as most fans on this message board) predict where teams will finish in the fall, why is it such a crime for BP to use PECOTA to attempt to do the same? If you disagree that's fine. It is difficult if not impossible for the average person to base predictions on the same amount of knowledge as PECOTA since we're talking about a gigantic database of information on baseball players throughout history, but certainly different analytical methods are not without value.

To the extent that there is misunderstanding with the projections it seems to center around the fact that what PECOTA is doing is projecting the most likely outcome based on the numbers throughout history (I am correct in assuming that there's absolutely no support behind the assertions last seasons individual performances/win totals are equally successful in projecting the current season, right?). Why people tend to assume that the PECOTA projections are meant to suggest it is extremely unlikely or impossible for the White Sox to win 81 games let alone 88 just because the projected win total is 80 is completely beyond me. When Peter Gammons picks the Twins to win the AL Central I'm sure he would not tell you he is absolutely positive in that outcome just as most of you aren't completely positive if you pick the Braves to win the NL East. It is an educated guess.

Several posters in this thread seem be attempting to argue against the PECOTA projections by stating that you have to play the games and that certain events cannot be forseen. That reality has certainly not been lost on anyone over at BP. The fact that PECOTA generates the potential for players to "breakout" and "collapse" is a means of further examining players' ability to have seasons that seemingly come out of nowhere. Furthermore the authors realize that sometimes events happen that absolutely no one plans for or sees coming, that's an obvious fact of life which is not limited to baseball. What I cannot understand is why some consider the fact that PECOTA is far from capable of perfectly predicting the future a reason to reject the system entirely and refuse to give the projections any consideration whatsoever. If I'm filling out a lottery ticket and you can tell me the first number, I'm not about to refuse to use that number just because you can't tell me the rest...


DaDawg has been out propellerheaded.

Lip Man 1
03-22-2005, 10:18 PM
Hey Jeremy's back!!!

The reason I mentioned the phythegorian hypotenuse or whatever the hell it's called is because they insisted there was no way the Twins could have won one division let alone three because the numbers didn't add up. I don't believe they ever picked the Twins to win anything (I could be wrong about this the 2nd or 3rd year) and yet they did.

Numbers aren't the sum of all knowledge. That's why they play the games on the field and not in some egghead's bedroom.

Lip

Lip Man 1
03-22-2005, 10:20 PM
Daver says:

"DaDawg has been out propellerheaded."

LOL I almost fell out of my chair reading that. :bandance:

Lip

Tragg
03-22-2005, 11:10 PM
What is with this love for the indians?


I think they hit their high water mark last year.

dugwood31
03-22-2005, 11:17 PM
80-82 sounds about right to me.

Garcia's an underachiever whose numbers were bad after the trade, El Duque is fragile, Contreras has never been effective, and as Prospectus put it, bragging about Garland being a 5th starter belies the fact that he pitches like a fifth starter.

Our bullpen is deeper. I feel good about it.

No one knows how Iguchi will do, Rowand had a breakout year last year but it was close to his ceiling. I expect him to dip a bit. Konerko had his career year last year -- I expect a dropoff. Pod might not get on base enough to create the spark that everyone thinks he'll create. Dye's likely to get hurt.

We went from having a monster middle of the order to a meidocre middle.
Sometimes I think posters on this board sound like Cubs fans with the unrealistic optimism. The best thing we have going for us is our weak division.

santo=dorf
03-23-2005, 12:01 AM
Garcia's an underachiever whose numbers were bad after the trade,.
*****.
His numbers were inflated due to USCF as it would happen to any other pitcher. He won 9 games with the Sox, had a road ERA of 3.25, a Road WHIP of 1.04, and a K/9 of 10.15 on the road with the Sox. Just terrible. :rolleyes:

El Duque is fragile.
So are Millwood, Sabathia, and Mays

Contreras has never been effective,.
Of course he has only been in the bigs for two years. What about 2003?

and as Prospectus put it, bragging about Garland being a 5th starter belies the fact that he pitches like a fifth starter.
I'm one of the biggest Garland bashers around, but even I can say he is probably the best 5th starter in the AL. If Garland was the 5th starter in 2003 or 2004, we'd be in the playoffs.

Rowand had a breakout year last year but it was close to his ceiling.
Based on what? :?:

santo=dorf
03-23-2005, 12:03 AM
EDIT: And Lip, why were you bashing the Pythagorean rankings? Last year they/it were/was off by one game, 2003 by two games, and 2002 it was off by five games...

You can manipulate the numbers as much as you want, but they still don't mean a damn thing. I would love to see Bill James explain where he got 1.83 from.

Banix12
03-23-2005, 12:09 AM
80-82 sounds about right to me.

Garcia's an underachiever whose numbers were bad after the trade, El Duque is fragile, Contreras has never been effective, and as Prospectus put it, bragging about Garland being a 5th starter belies the fact that he pitches like a fifth starter.

Our bullpen is deeper. I feel good about it.

No one knows how Iguchi will do, Rowand had a breakout year last year but it was close to his ceiling. I expect him to dip a bit. Konerko had his career year last year -- I expect a dropoff. Pod might not get on base enough to create the spark that everyone thinks he'll create. Dye's likely to get hurt.

We went from having a monster middle of the order to a meidocre middle.
Sometimes I think posters on this board sound like Cubs fans with the unrealistic optimism. The best thing we have going for us is our weak division.


Kind of a negative assessment.

Garcia is an underacheiver, but he's still pretty darn good for an underacheiver. El Duque has had the one injury, albeit a big one, and came back a bit too quick last season and wore down. Maybe during the offseason he gained a bit of strength, the season will dictate that. Contreras is very effective 2/3rds of the time, though the other 1/3 he stinks. Garland is at least consistant and didn't have great run support most of last season.

It's hard to say if Rowand has peaked, since he didn't even get all the at-bats he would have last season because he was on the bench the first two months to get at-bats for Willie harris while Uribe was on his hot streak.

Dye has an injury history, no doubt about it, but he will benefit offensively from not playing most of his games in Oakland.

If Rowand gets all his playing time and Dye plays most of the season, they should at least come close to the annual power output of Lee and Maggs. Pierzynski and Iguchi should make up the difference and then some.

Konerko might have peaked, but even with a slight dropoff he might be the best slugging 1b in the American league, with possible exceptions being Sexson if healthy, Texiera and Giambi if he's back on the juice.

Pods, like Rowand and Konerko, had two drastically different years between 03 and 04. Just the other way around where Pods struggled last season instead succeeding. In Milwaukee there was pressure on him to be the whole offense with Jenkins often injured and nobody else really hitting than overbay. Hopefully he succeeds knowing he doesn't have to steal home every game to win games and is no longer needed for power hitting.

I'm not saying that they will definitely win over 90 games, a .500 record sounds like a decent conservative estimate of the sox this season. But we basically did that last season missing two power hitters for last half of the season, forfeiting every fifth start as a loss because of our four man rotation + whipping boy and with a bullpen that started the season with Billy Koch at closer and Mike Jackson and Jon Adkins setting up.

So there is reason for optimisim, because this team at least from the start of this season seems more complete than last years which finished around .500. Do things have to go right for the sox to succeed. Yes but that is true about any team, save maybe the yankees.

batmanZoSo
03-23-2005, 12:24 AM
However, PECOTA does have a track record of capturing mean (50% probabilty) performances of players. Of course, when players move to 80-90% or down to 10-20%, then different outcomes occur.

Read the attached article on see how PECOTA translates into projected wins--84 wins last year, down to 78 as Mags departs, back to 82 with the signing of Dye, down to 77 with the infamous Lee-Podsednik trade, ticks up 78 with the signing of Hernandez, up to 80 with the signing of Pierzyntski, between 80-81 with the signing of Iguichi.

It is a methodology which people are free to accept or reject. But it is a methodology based on proven outcomes and is better starting point for discussion than "I just don't buy that".

PECOTA sucks and so do we at predicting things. I wish it would all stop...

dugwood31
03-23-2005, 12:36 AM
*****.
His numbers were inflated due to USCF as it would happen to any other pitcher. He won 9 games with the Sox, had a road ERA of 3.25, a Road WHIP of 1.04, and a K/9 of 10.15 on the road with the Sox. Just terrible.



His ERA last year with the Sox was 4.46. Slightly better than the league average, but 3rd starter territory. His numbers were inflated by USCF, but he does have to pitch there. I'm too lazy to look up the home/road splits :cool: but if his Sox ERA was 4.46, his road ERA with the Sox was 3.25, then his home ERA with us was disastrous. In '02 and '03 his ERA was below league average -- in a pitchers park. He improved in '04 -- his contract year. What happened after he signed that deal?

So are Millwood, Sabathia, and Mays

So aren't Santana and Radke! And Mays is penciled in as a 5th starter. El Duque is our 3rd. Hernandez has thrown 325 innings in the last 4 years! That's our $5 million third starter people.

Of course he (Contreras) has only been in the bigs for two years. What about 2003?

He started 9 games in '03. His ERA was 5.50 last year! Only blind optimism suggests he'll be good this year. Although I understand his upside, counting on it is problematic.

I'm one of the biggest Garland bashers around, but even I can say he is probably the best 5th starter in the AL. If Garland was the 5th starter in 2003 or 2004, we'd be in the playoffs.


One problem: Garland is better than Contreras, and Hernandez will get hurt, so he's really not our number 5, unless we get lucky. Also, I'd take Bronson Arroyo and Paul Byrd in our rotation, and they're both 5th starters.

Rowand was never a can't-miss type prospect, which would suggest lower upside. His work ethic, short swing and overall approach got him to his peak sooner than other young players. Don't get me wrong, I love Aaron and expect a good year, but he won't be 7th in AL in slugging this year. His .900 OPS will settle in around .850, and I'll love him for it.

I just think we're overvaluing this whole idea that grinderball will solve our problems. We'll be good if Crede breaks out, if Uribe doesn't slip, if Iguchi is decent, if Podsednik gets on base more, if Dye stays healthy, if Frank gets healthy, if El Duque stays healthy, if Contreras magically figures it out, if Garcia pitches like he's capable of, if Buehrle's stress reaction is really just that, if Shingo continues to fool people.

Obviously we can be good if a few of the above don't happen, but we need most of them to happen to even have a chance. Thank goodness the Twins have a smaller budget than we do!

mdep524
03-23-2005, 01:31 AM
I'm all for expanding the box score and using new tools to evaluate players. most of it is useful information, especially when evaluating individual players. Statistics shouldn't be the only tool, but it is an incredibly useful tool

However using this info to try and accurately predict the outcome of a team over an entire season is a waste of time. Too many intangibles that can't be taken into account.

Also, I don't want anyone to find a way to accurately project the season before it's played, that would make watching baseball complete waste of time. The mystery of what will happen is part of the mystique. Best. Post. Ever.

Stats are great, fun and interesting, and are an undeniable part of baseball. But nothing in baseball is fool proof: neither stats, Ben Davis' statistically-immeasurable defense, nor Ozzie Guillen's effect on team chemistry is going to accurately predict an outcome. And as Banix said, this is a good thing! That's why they play the games!

Cowch44
03-23-2005, 01:44 AM
With wins being based on the players..How many wins are they predicting for the Yankees?:?:

fquaye149
03-23-2005, 05:15 AM
So aren't Santana and Radke! And Mays is penciled in as a 5th starter. El Duque is our 3rd. Hernandez has thrown 325 innings in the last 4 years! That's our $5 million third starter people.



One problem: Garland is better than Contreras, and Hernandez will get hurt, so he's really not our number 5, unless we get lucky. Also, I'd take Bronson Arroyo and Paul Byrd in our rotation, and they're both 5th starters.

Rowand was never a can't-miss type prospect, which would suggest lower upside. His work ethic, short swing and overall approach got him to his peak sooner than other young players. Don't get me wrong, I love Aaron and expect a good year, but he won't be 7th in AL in slugging this year. His .900 OPS will settle in around .850, and I'll love him for it.

I just think we're overvaluing this whole idea that grinderball will solve our problems. We'll be good if Crede breaks out, if Uribe doesn't slip, if Iguchi is decent, if Podsednik gets on base more, if Dye stays healthy, if Frank gets healthy, if El Duque stays healthy, if Contreras magically figures it out, if Garcia pitches like he's capable of, if Buehrle's stress reaction is really just that, if Shingo continues to fool people.

Obviously we can be good if a few of the above don't happen, but we need most of them to happen to even have a chance. Thank goodness the Twins have a smaller budget than we do!

a.) radke and santana have very little to do with whether we finish 2nd or 3rd.

b.) Joe Mays may be PENCILED in at 5th by gardenhire, but any sort of logic would dictate that he is a 3rd ahead of silva and lohse.

c.)YOU'D take Arroyo and Paul Byrd, but that's because you have the "they're not on the sox" colored glasses on. Last I checked, Paul Byrd's not in the AL, and besides one good year in KC his numbers are not at all better than garlands, ERA or Win. Not to mention if you want to talk about injury prone. Arroyo? Well his career ERA's 4.60...comparable to Garland, but he's averaged less than 100 innings per year and less than 5 wins. Do you really know what you're going to get with him? But he plays with another team so we assume there are no question marks with him, right? Only GARLAND with his "bad attitude". forget the actual NUMBERS.

d.) can't miss prospects have nothing to do with upside. It has to do with probability to REACH that upside. For instance: Jerry Owens is NOT a can't miss prospect. However his ceiling is much higher than, say, Josh Fields'. However Fields is more likely to fulfill his potential. I don't know exactly how Rowand fits into this, but can't miss has little to do with ceiling.

e.) put away your strawman of grinderball. This team sacrificed very little power, and will still hit plenty of 3-run homers. Not to mention I have heard very few people on this board champion the grinder aspect. It will be nice to see, but PITCHING is where this team has improved and you damn well know it.


If you want to approach our team like it has so many more question marks than any other team (especially the bleeping indians!) make sure you at least get your pessimistic facts straight.

Ol' No. 2
03-23-2005, 09:21 AM
But the people who talk about heart and guts can have absolute faith in measuring those "intangible" skills?

Please. Like Hawk today saying he doesn't care about Joe Crede's .239 average because he's got the guts to get some big hits when it counts. If he sucks the other 97% of the time, it is a big deal.

They're just using stats to make a prediction. Nobody, nobody, is saying it's the be all end all of predictions. Unless, of course, it said the Sox would win it all. Then we'd love them.Just because you can't measure intangibles (by definition), that doesn't mean they don't exist. However, I would agree that they are often overblown and used as an excuse to boost a player that doesn't deserve it.

Statistical analysis is a useful tool, but unfortunately, too often misused by people who don't understand statistics. PECOTA projections are a case in point. Anyone with a calculator can do a simple regression and assign "values" to players and add them up to see how many wins a team will have. Unfortunately, it's BS. All you're calculating is the mean expected value. But when there's a large variation of expected values, the mean is almost worthless. What good is projecting 84 wins if its error range is +/- 10 games?

Flight #24
03-23-2005, 09:46 AM
His ERA last year with the Sox was 4.46. Slightly better than the league average, but 3rd starter territory. His numbers were inflated by USCF, but he does have to pitch there. I'm too lazy to look up the home/road splits :cool: but if his Sox ERA was 4.46, his road ERA with the Sox was 3.25, then his home ERA with us was disastrous. In '02 and '03 his ERA was below league average -- in a pitchers park. He improved in '04 -- his contract year. What happened after he signed that deal?




The fallacy in your argument is that yes, Freddy has to pitch there, but the park will have similar effects on the opposing pitcher. But because the opposing pitcher varies from player to player, the impact on their averages is less then it is on his. Just picking a couple of players: Brad Radke has a 5.16 EA over the past 3 years at USCF (5 starts). Johan Santana is at 4.88 (3 starts). The point in all of this is that if Freddy Garcia starts in those 11 games against these pitchers and puts up a high home ERA, chances are that the opposing pitchers have one equally high or higher. They just spread it out over more players. It's a problem with the use of averages, and it's the reason why park factors, and various methods of predicting "independent ERAs" have been developed.

And that's the reason why a quick & dirty method of making an accurate comparison between pitchers is to look at road stats, which are pretty consistent across players within a league.

Ol' No. 2
03-23-2005, 09:49 AM
The fallacy in your argument is that yes, Freddy has to pitch there, but the park will have similar effects on the opposing pitcher. But because the opposing pitcher varies from player to player, the impact on their averages is less then it is on his. Just picking a couple of players: Brad Radke has a 5.16 EA over the past 3 years at USCF (5 starts). Johan Santana is at 4.88 (3 starts). The point in all of this is that if Freddy Garcia starts in those 11 games against these pitchers and puts up a high home ERA, chances are that the opposing pitchers have one equally high or higher. They just spread it out over more players. It's a problem with the use of averages, and it's the reason why park factors, and various methods of predicting "independent ERAs" have been developed.

And that's the reason why a quick & dirty method of making an accurate comparison between pitchers is to look at road stats, which are pretty consistent across players within a league.This is why ERA+ was invented. It's probably the most meaningful.

Iwritecode
03-23-2005, 10:17 AM
This is why ERA+ was invented. It's probably the most meaningful.

Now in another thread you said that wins and losses were the most important statistic when evaluating pitchers... :tongue:

Flight #24
03-23-2005, 10:17 AM
This is why ERA+ was invented. It's probably the most meaningful.

Yup. And for comparison, Freddy's career ERA+ of 114 is comparable to guys like Colon (116), Mulder (115), and Radke (114). But I suppose those guys are all #3 starters too.

Lip Man 1
03-23-2005, 11:50 AM
Fquay:

Byrd is with the Angels I believe.

Lip

wdelaney72
03-23-2005, 12:21 PM
Impressive, since he's dead.
http://img.photobucket.com/albums/v311/wdelaney72/RachelPhelps.jpg "Cross him off, then."

dugwood31
03-23-2005, 12:34 PM
YOU'D take Arroyo and Paul Byrd, but that's because you have the "they're not on the sox" colored glasses on. Last I checked, Paul Byrd's not in the AL



Paul Byrd is on the Los Angeles Angels of Anaheim. Last I checked, they're in the AL. Byrd missed '03, but was good in '02 and good last year when he came back. Comparabe to El Duque, who's our number 3.

dugwood31
03-23-2005, 12:40 PM
e.) put away your strawman of grinderball. This team sacrificed very little power, and will still hit plenty of 3-run homers. Not to mention I have heard very few people on this board champion the grinder aspect. It will be nice to see, but PITCHING is where this team has improved and you damn well know it.

Pitching will be the factor if we win. I like our bullpen, which means I'm not pessimistic for pessimism's sake. And I agree that our pitching could be decent, but it would take a lot of if's. Grinderball is not a straw man, most of the optimism I hear comes from the fact that people think speed will overcome the loss of a potentially healthy Maggs, CLee and yes, even Valentin. I hope it does, but I'm not convinced.manufacture more runs this year, but we also will hit fewer bombs.

MeanFish
03-23-2005, 12:42 PM
Now in another thread you said that wins and losses were the most important statistic when evaluating pitchers... :tongue:static String mostImportantStatIMHO(String theStat){
if(theStat == "ERA" || theStat == "ERA+"){
return "ERA+";
}
else{
return "Win/Loss Ratio";
}
}

I'm a dork. And the forum parses out leading whitespace. Grr.

fquaye149
03-23-2005, 01:50 PM
Fquay:

Byrd is with the Angels I believe.

Lip


eep! :redface: who can i blame this on?

I blame the schools!

(no offense, hal)

fquaye149
03-23-2005, 01:51 PM
Pitching will be the factor if we win. I like our bullpen, which means I'm not pessimistic for pessimism's sake. And I agree that our pitching could be decent, but it would take a lot of if's. Grinderball is not a straw man, most of the optimism I hear comes from the fact that people think speed will overcome the loss of a potentially healthy Maggs, CLee and yes, even Valentin. I hope it does, but I'm not convinced.manufacture more runs this year, but we also will hit fewer bombs.

Where are you hearing it? Certainly not on this board.

Oh you mean articles in paper's like the tribune

my mistake, it's not a strawman at all.

Iwritecode
03-23-2005, 03:12 PM
manufacture more runs this year, but we also will hit fewer bombs.

Yeah, leading the league in homeruns last year helped soo much!!!

Banix12
03-23-2005, 04:23 PM
Grinderball is not a straw man, most of the optimism I hear comes from the fact that people think speed will overcome the loss of a potentially healthy Maggs, CLee and yes, even Valentin. I hope it does, but I'm not convinced.manufacture more runs this year, but we also will hit fewer bombs.

We led the league in HR's last season without Maggs and Thomas for most of the season. As a team the sox are still going to hit a ton of Homeruns, the only difference this year is the sox should have the option of manufacturing runs as well, something we couldn't do last year.

If I had to guess what the net loss in sox Homeruns from last year to this year will be, I'd say somewhere around -30 as a conservative estimate. We have more power from the catcher position and 2b with Pierzynski and Iguchi. A wash at SS with Uribe taking over for Valentin (do you think Valentin has another 30 HR season in him? I don't) Rowand should get 500+ at bats and could be a 30 HR guy, at worst hitting in the mid 20's. Dye should hit in the mid-20's or better. Everett seems to be back to his old self, and Thomas is coming back during the season.

We still have a ton of power, more than most of the league, and we have more speed at the top of the lineup, which is something we have been missing for years. So it's not that the sox chose Speed over Power throughout the whole lineup, we just augmented the lineup so we have power with speed.

dugwood31
03-23-2005, 09:54 PM
Yeah, leading the league in homeruns last year helped soo much!!!

We were second to last in team ERA last year and still finished above .500. Offense wasn't the problem.

dugwood31
03-23-2005, 10:00 PM
Where are you hearing it? Certainly not on this board.
What I'm hearing on this board is an overly optimistic assessment of our team. The only way to justify this optimism is to be convinced our starting pitching has improved by leaps and bounds, and that our new, speedier offense will be more consistent. Perhaps using the term grinderball was offensive, but that's what I understand it to mean in the context of the 05 Sox.

I think we're about a .500 team. I'm surprised that it's so controversial around here.

dugwood31
03-23-2005, 10:04 PM
If I had to guess what the net loss in sox Homeruns from last year to this year will be, I'd say somewhere around -30 as a conservative estimate. We have more power from the catcher position and 2b with Pierzynski and Iguchi. A wash at SS with Uribe taking over for Valentin (do you think Valentin has another 30 HR season in him? I don't)

We're absolutely better offensively at catcher. Given that Jose hit 25+ for 5 straight years, and Uribe's never hit 25, I'm not sure you can say it's a wash power-wise. Uribe's younger and might turn out to be twice the player, but in terms of homers we'll be lucky if it's a wash. I hope you're right about Iguchi. We'll see. My instinct tells me that if he puts up the kind of numbers some think he will, more teams woulda taken a flier at $2.5 mil, especially given the weakness at 2B around the league. Hope I'm wrong.

Daver
03-23-2005, 10:13 PM
The propellerhead has spoken.

fquaye149
03-24-2005, 03:01 AM
that would be a fun tag to have^^^

DocWolf
03-30-2005, 05:04 PM
games are played on the field and not in a computer. This club has some uncertainties, (Contreras?) but what AL central team doesn't, especially the Twinkies. :hamms