BBTF's Newsblog Discussion :: THT: Outsmarting Pythagoras

Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Monday, September 17, 2007

THT: Outsmarting Pythagoras

Sal Baxamusa says!

As an example, let’s look at what happened in the AL West. Here are the standings since Sheehan’s column ran:

On July 19, 2007
Team W L pW pL RS RA
Seattle 53 39 47.0 45.0 456 445
Oakland 45 50 48.7 46.3 402 391

Since July 19, 2007
Team W L pW pL RS RA
Seattle 25 31 24.3 31.7 270 311
Oakland 29 27 27.0 29.0 299 308

Wouldn’t you know it? The A’s have gained four real games on the Mariners since then and have played almost three games better in the Pythagorean world. Furthermore, Seattle’s record is just under its Pythagorean projection. Whatever enabled the Mariners to outdo their Pythagorean record earlier in the year seems to have worn off in mid-July, despite Sheehan’s observation that “the Mariners are the better team and will continue to be so.”

It’s time to stop trying to outsmart ourselves. While we may be able to point to certain factors that have caused a team to have a large Pythagorean differential, it’s folly to use those facts as evidence that the trend will continue.

AuntBea Posted: September 17, 2007 at 09:55 PM | 21 comment(s)
Related News: General, Sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 1 of 1 pages

1. Sid Hārtman Gautama Posted: September 17, 2007 at 11:42 PM (#2529552)

Before people use this as an excuse to pile on Joe Sheehan (since for some reason BPro draws the ire of primates):

NOTHING in what I wrote should be construed as an attack on Joe Sheehan, either personally or analytically. I used his article as an example for the reason that Joe is a leading figure in analytical circles (and with good reason). I've heard from some that this was an attack on Joe. Maybe I wasn't clear enough, but NOTHING in there was meant as a slight against Joe (whom I hold in high regard).

I know there's a lot of sensitivity about "anonymous" internet writers attacking other writers, and I don't intend to be part of that breed. At the same time, I'm not going to be refrain from directly disagreeing with other writers if I do so in a collegial spirit. If I failed to communicate that clearly - my bad. But it certainly wasn't my intent attack Joe.

Now let's talk about Pythagoras, okay?

2. AuntBea Posted: September 18, 2007 at 08:11 AM (#2529678)

Sal, I'm with you on this one. The two biggest mistakes people make is to:

1) Overweight recent results and
2) Search for patterns in the void

Both errors lead to disrespectin' Pythagoras.

Regarding the D-backs specifically, we heard a lot of justification of their performance on this site as well.

3. Sid Hārtman Gautama Posted: September 18, 2007 at 01:54 PM (#2530007)

Indeed, it was Rosenthal's column on the DBacks that got me riled up enough to write this in the first place.

4. Greg Maddux School of Reflexive Profanity Posted: September 18, 2007 at 02:10 PM (#2530023)

Joe is a leading figure in analytical circles

Good one.

5. With 17th Pick, From LA, 1k5v3L KcoLLoP Posted: September 18, 2007 at 02:11 PM (#2530024)

Maybe you meant "Joe is a leading figure in analyzing circles"?

6. Sid Hārtman Gautama Posted: September 18, 2007 at 02:30 PM (#2530041)

Good one.

Maybe you meant "Joe is a leading figure in analyzing circles"?

Christ, did anyone read the first comment. Sheesh.

7. Greg Maddux School of Reflexive Profanity Posted: September 18, 2007 at 02:37 PM (#2530046)

Yes, the portion I quoted came from it.

8. With 17th Pick, From LA, 1k5v3L KcoLLoP Posted: September 18, 2007 at 02:49 PM (#2530057)

I read Maddog's comment which quoted the first comment. The real question, Sal, is where Rosenthal ranks in those analytical circles...

Btw, Joe would be a better analyst if he weren't too busy inventing completely random excuses for jumping on and off team bandwagons...

9. Tim Lincecum doesn't Wang Chung tonite (GGC) Posted: September 18, 2007 at 02:52 PM (#2530060)

I always thought that Sheehan:BPro::Thorn:Palmer or Aaron:Moses

10. Tim Lincecum doesn't Wang Chung tonite (GGC) Posted: September 18, 2007 at 03:22 PM (#2530092)

SAT like anologies have killed this thread.

11. Sid Hārtman Gautama Posted: September 18, 2007 at 03:27 PM (#2530096)

SAT analogies:BTF threads::_____:babies

12. TDF, situational idiot Posted: September 18, 2007 at 03:31 PM (#2530100)

I guess I'm missing something, and my confusion may be what some of these writers are expressing: Just because Seattle overperformed vs. Pythag early in the season, why would you assume they'll regress back to the norm? Why wouldn't you assume that they'll "win to" their Pythag projection the rest of the way out?

13. With 17th Pick, From LA, 1k5v3L KcoLLoP Posted: September 18, 2007 at 03:33 PM (#2530104)

Why wouldn't you assume that they'll "win to" their Pythag projection the rest of the way out?

That's more or less what has happened in AZ, really. The M's probably "underperformed" their pythag projection the rest of the way because of the bullpen explosion and their horrendous manager, and some bad luck to boot.

14. Sid Hārtman Gautama Posted: September 18, 2007 at 03:33 PM (#2530105)

I guess I'm missing something, and my confusion may be what some of these writers are expressing: Just because Seattle overperformed vs. Pythag early in the season, why would you assume they'll regress back to the norm? Why wouldn't you assume that they'll "win to" their Pythag projection the rest of the way out?

15. AuntBea Posted: September 18, 2007 at 03:49 PM (#2530123)

Why wouldn't you assume that they'll "win to" their Pythag projection the rest of the way out?

In general, you would assume that. Ans as the quoted excerpt shows, that is also what has happened.

16. AuntBea Posted: September 18, 2007 at 03:51 PM (#2530126)

Maybe it's my fault, I didn't bother to format the excerpt concerning the Mariners.

17. TDF, situational idiot Posted: September 18, 2007 at 04:32 PM (#2530171)

I didn't suggest they would regress (or didn't mean to suggest it anyway). What I'm getting at is that the way a team will play in the future is best guessed at by using Pythag and how many runs you think they will score/allow (not what they've already done).

I'm not saying you in particular said it here, but that seems to be the statement people make - "The Reds are 8 games over their Pythag record; they're due for a fall". I've always taken that to mean that they cannot stay +8 no matter what, that what they've already done predicts what they will do pretty closely.

18. AuntBea Posted: September 18, 2007 at 04:45 PM (#2530187)

I've always taken that to mean that they cannot stay +8 no matter what, that what they've already done predicts what they will do pretty closely.

Actually, if you are +8 through the first half of the season, and "closely match" pythagoras the second half, you will still be close to +8 by the end of the season. However, technically this is a regression to the mean, as the longer the season, the higher the standard deviation for the difference between pythagoran and actual wins.

Imagine the season goes 1 million games after the first 80. If you exactly match pythagoras during those million games, you end up +8. +8 after a million games is essentially an exact match.

19. TDF, situational idiot Posted: September 18, 2007 at 04:58 PM (#2530205)

Actually, if you are +8 through the first half of the season, and "closely match" pythagoras the second half, you will still be close to +8 by the end of the season. However, technically this is a regression to the mean, as the longer the season, the higher the standard deviation for the difference between pythagoran and actual wins.

I think you're missing my point - many (most?) would say that the Reds, after being +8 in the 1st half, would approach -8 in the second. My point (which you seem to agree to) is that that's faulty logic.

20. AuntBea Posted: September 18, 2007 at 05:02 PM (#2530209)

Ok. I see your point now. In the Ankiel thread, someone said that his latest 2 for 28 was a "regression to the mean", implying that it was overdue to happen. As you state, this faulty logic is all too common.

21. Sid Hārtman Gautama Posted: September 18, 2007 at 08:57 PM (#2530618)

In the Ankiel thread, someone said that his latest 2 for 28 was a "regression to the mean", implying that it was overdue to happen.

Right. That type of logic is as faulty as saying that a hot streak will continue.

Page 1 of 1 pages

You must be Registered and Logged In to post comments.

<< Back to main

Monday, September 17, 2007

THT: Outsmarting Pythagoras

Reader Comments and Retorts

Support BBTF

My Bookmarks

Hot Topics