Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Saturday, March 10, 2007

Statistical program predicts aging White Sox will win only 72 games (RR)

Natch…marginal lineup valueless to team.

Baseball Prospectus, considered the new-age statistical bible, projects the White Sox to finish with a 72-90 record this season.

“Well, we’re screwed now,” team captain Paul Konerko said with a sarcastic laugh. “I guess we’ll just have to battle through.”

...Naturally, the White Sox scoff at this, although they could use it to help restore the massive chip on their shoulder that was lost after the 2005 season.

“That’s a good sign for us because usually they’re wrong about everything regarding our dealings,” general manager Ken Williams said. “What can you do? We put the best team together we can, and we think we’re going to end up somewhere in the mid-90s, although there are all kinds of variables off that.

Repoz Posted: March 10, 2007 at 11:00 PM | 35 comment(s) Login to Bookmark
  Tags: sabermetrics, white sox

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.

   1. bibigon Posted: March 10, 2007 at 11:35 PM (#2310060)
This is actually a pretty good article - Silver comes off pretty well in particular. Are these projections on BPro somewhere by the way? I haven't seen them yet.
   2. Shock Posted: March 10, 2007 at 11:40 PM (#2310061)
Why do people keep using the term "new-age" to refer to sabrmetrics? Isn't statsitcal analysis basically, um, the opposite of "new-age" ####? My mother is into new-age stuff, and it's all about angels and aura and mystique and all that hubbub. Hardly Sabrmetrics.
   3. Dag Nabbit at ExactlyAsOld.com Posted: March 10, 2007 at 11:57 PM (#2310066)
Repoz - the url doesn't work. Keeps saying "page not found" for me.
   4. Darren Posted: March 11, 2007 at 12:04 AM (#2310069)
Aha. The missing ingredient in the 2007 Chicago White Sox is... Love? Alright, who's been messing around with the PECOTAnator 5000?
   5. Greg Maddux School of Reflexive Profanity Posted: March 11, 2007 at 12:06 AM (#2310070)
The computer didn't like Darin Erstad because <strike>of his age (32) and injury history</strike> he sucks and didn't like expected fifth starter Gavin Floyd because <strike>of his inexperience</strike> he also sucks.

I mean, really? You're going with "inexperience" with those 7.00+ ERAs staring you in the face?
   6. Walt Davis Posted: March 11, 2007 at 12:09 AM (#2310071)
that's weird, worked 5 minutes ago.

I find a mean projection of 72 pretty hard to swallow. That almost puts 100 losses into the "not that surprising" category and a lot would have to go wrong for the Sox to lose 100 games.
   7. The Anthony Kennedy of BBTF (Scott) Posted: March 11, 2007 at 12:14 AM (#2310075)
yeah. i don't see it happening. injuries could put them in the low 80's range, but i'd be surprised if they're a sub .500 team.
   8. alkeiper Posted: March 11, 2007 at 12:16 AM (#2310077)
Joe Sheehan appeared on ESPNNews on Tuesday, and he disagreed with the projection as well. Obviously there are a couple factors it can't detect.

72 wins would equal a .444 winning percentage. That would be the Sox' worst season since 1989.
   9. bibigon Posted: March 11, 2007 at 12:17 AM (#2310078)
I find a mean projection of 72 pretty hard to swallow. That almost puts 100 losses into the "not that surprising" category and a lot would have to go wrong for the Sox to lose 100 games.


It seems to mainly come from the White Sox being projected by PECOTA to have a really bad starting staff, with ERAs ranging from 4.50(Vazquez) to 6.56 (Floyd). None of the offensive projections seem hugely out whack, just lots of regression to the mean.

Overall, I'd easily take the over on them for 72 wins, but I'm not sure I'd take the over on like 81 wins.
   10. Joyful Calculus Instructor Posted: March 11, 2007 at 12:30 AM (#2310080)
The projections for the 99-victory 2005 world champions were 71-91 and a fourth-place finish


In other words, teh Sox pwn computer projections.
   11. scareduck Posted: March 11, 2007 at 12:37 AM (#2310083)
My mother is into new-age stuff,

Rhymes with "sewage" ...
and it's all about angels

... it's reflexive -- the Angels don't believe in OBP, either...
and aura and mystique

... Curt Schilling's nightclub dancers. It's all part of the Great Circle of Life. Or Outs. Or something.
   12. Danny Posted: March 11, 2007 at 12:52 AM (#2310088)
The projections for the 99-victory 2005 world champions were 71-91 and a fourth-place finish

Where did that come from? PECOTA had the Sox at 80-82 for 2005.

For 2007, PECOTA has them scoring 778 runs and allowing 868 runs. That's more runs than the Sox scored in 2005, so I guess it's not that unreasonable. But the 868 runs is 70 more than they allowed in 2005. It's not just that PECOTA thinks Floyd is going to be terrible, while Graland and Contreras will be below average. The Prospectus depth charts are also projecting them to get over 300 IP from Danks, Haeger, Tracey, Logan, and Masset--who are all projected to be varying degrees of awful.
   13. Dr. Vaux Posted: March 11, 2007 at 01:05 AM (#2310090)
I think their pitching will be awful, but their offense will score tons of runs and they'll win the division. But that's not based on anything but my patented pessemistic gut.
   14. Dag Nabbit at ExactlyAsOld.com Posted: March 11, 2007 at 01:09 AM (#2310091)
Going by SG's annual reseason simulations blow out, the White Sox have exceeded projections every year under Guillen.
   15. Greg Maddux School of Reflexive Profanity Posted: March 11, 2007 at 01:13 AM (#2310092)
Where did that come from? PECOTA had the Sox at 80-82 for 2005.

From the brilliant mind of Joe Sheehan. The White Sox had underperformed their projection in previous years, and, much like a fan of the Angels, he decided that should be applied to all future projections.
   16. user Posted: March 11, 2007 at 01:15 AM (#2310095)
PECOTA would rather give jose lima a rotation slot than gavin floyd.
   17. Rich Rifkin I Posted: March 11, 2007 at 03:16 AM (#2310119)
Gavin Floy'd mother would rather give jose lima a rotation slot than gavin floyd.
   18. shoewizard Posted: March 11, 2007 at 03:17 AM (#2310156)
If you go to BP Fantasy Link on their front page, and then go to Depth Charts and pick the white sox from the drop down menu, you will find player projections with playing time doled out, and the teams projected won loss, runs scored/allowed, and team batting line.

It's subscriber content.
   19. Jorge Luis Bourjos (Walewander) Posted: March 11, 2007 at 03:45 AM (#2310135)
Not 72, but I'm more worried about the Indians. I think the Tigers, Twins, and Sox will all regress, partly due to an improved KC squad. OTOH, I don't think Cleveland has the pitching to win the division, so I'm left with the 4-team tossup everyone else has.
   20. Walt Davis Posted: March 11, 2007 at 03:49 AM (#2310163)
You never know with pitchers, but in 2006, the Sox rotation gave them 159 starts and Contreras was the low guy with 196 IP. In 2005, it was 152 and again over 1000 innings total. If the Sox once again get 1000 innings from their rotation (or even 800+ from their top 4), they'll be doing just fine. That will also help them hide whatever flaws there might be in their bullpen.

I have no doubt there are scenarios where the Sox win just 72. A mean projection around 500 wouldn't surprise me given that division. But dropping 90 runs on offense and giving up 70 more on defense vs. 2006 ... even Erstad and Floyd aren't that bad.
   21. Bring Me the Head of Alfredo Griffin (Vlad) Posted: March 11, 2007 at 03:50 AM (#2310140)
Go away, Rob.

I can almost see 72 wins for an injury-wracked Sox team, since the division looks pretty tough.
   22. The Anthony Kennedy of BBTF (Scott) Posted: March 11, 2007 at 08:56 AM (#2310194)
rob, go waste your time at xoxohth.com

they're your sort of people.
   23. Gaelan Posted: March 11, 2007 at 10:51 AM (#2310199)
Is Rob the worst person on the planet. I think so.
   24. Hang down your head, Tom Foley Posted: March 11, 2007 at 03:59 PM (#2310231)
What did Rob say?
   25. shoewizard Posted: March 11, 2007 at 04:36 PM (#2310241)
Not worth repeating. He had an offensive ID, and made some offensive comments. Just trolling.
   26. 1k5v3L Posted: March 11, 2007 at 04:41 PM (#2310244)
The White Sox really need another solid outfielder. Even if Brian Anderson plays better, and Dye hits the average of his last two season, the Sox still have a desperate need for more production out of their third outfielder...

And maybe it's just me, but the Freddy Garcia trade will make Kenny W look really stupid really soon...
   27. sydhe Posted: March 11, 2007 at 09:17 PM (#2310369)
On the bright side, they'll still finish 15 games ahead of the Royals.
   28. K-BAR, J-BAR (trhn) Posted: March 11, 2007 at 10:08 PM (#2310404)
The problem with the White Sox win total is less in the projections themselves, but more in the assumptions about how the Sox would respond to those projected performances. There's no way Gavin Floyd would be allowed to start 21 games if his ERA were over 6.50. Danks or Haeger would probably be put into the 5th spot after about 5-10 bad Floyd starts. And if the Sox still were in striking distance for the playoffs after 15 or so bad starts from the 5th starter spot, Kenny would make a deal for another starter. Offensively, if Fields (775+ OPS), Sweeney (750+ OPS) and Erstad (~600 OPS) matched their PECOTA projections, Erstad wouldn't be on the field enough to put up a -13 VORP. The assumption that the Sox would irrationally deploy the talent they have on hand is probably costing them 3-4 wins. While a 76 win projection seems a bit low, it's more reasonable to me than 72.

The problem I have with this particular projection is that it will wind up a tout line if the White Sox completely collapse. If PECOTA's wrong, they can say that no one ever believed it. If it's right, odds are there will be a line somewhere next year about how PECOTA was dead on. Call it the "successfully predicted a break out for MarK Bellhorn" syndrome. Prior to Bellhorn's big season with the Cubs, his DT was terrible but the player comment was pretty positive. The next year there was a sentence about it on the back of the book. While I don't think they should be slavishly devoted to PECOTA or DTs, it frustrates me that Prospectus has a tendency to play both sides. Slavish devotion to PECOTA would at least make their results falsifiable, to the extent one can falsify a system that provides a broad range of percentile projections.
   29. Walt Davis Posted: March 12, 2007 at 02:40 AM (#2310535)
The problem with the White Sox win total is less in the projections themselves, but more in the assumptions about how the Sox would respond to those projected performances.

This is the problem with all projection systems and it's unsolvable. You might be able to tinker, running more simulations under differing assumptions about playing time. But most of the things you suggest, especially the mid-season trade, aren't able to be modelled. Would Williams go out and get someone? Teams often fail to fill that need -- e.g. last year lots of folks predicted the Sox would pick up a LF or CF. And who would he get? Who would be available? What teams would be out of the playoffs and selling players? How much would he have to give up? And of course if they're not in contention, he won't be picking up anybody but probably trading away Buehrle.

You're right they won't give Floyd 21 starts if he posts that kind of ERA. Just like the Yanks last year wouldn't give Chacon more than 11 starts with his 7.00 ERA. Only problem is they replaced him with Ponson (3 starts, 10.47), Small (3, 8.46), Wilson (1, 8.64), Lidle (9, 5.16) before getting some decent starts out of a couple kids. So it's true he won't get to suck that badly but it's far from guaranteed that the collection of guys who replace him will be better.

Also, if say two of the Sox top 4 starters get hurt (a real possibility for any staff) or if they aren't i contention and trade one, there's a very good chance Floyd gets at least 21 starts even with a 7.00 ERA.

So no projection system will do all that well -- including Kenny Williams' projection system. I'm gonna go out on a limb and say that he didn't project the 2006 Sox to give up 4.9 r/g (10th in league). If he did know they were gonna be that bad, he really did a lousy job as GM. :-)
   30. K-BAR, J-BAR (trhn) Posted: March 12, 2007 at 07:22 AM (#2310610)
For the most part, I agree with your points, Walt. One thing I'd like to point out is that in the case of the White Sox, PECOTA--or rather BPro's playing time forecaster thingy--projects them to act stupidly in the face of better options. Haeger & Danks project to have ERAs 1 run better. Obviously those projections are highly volatile, but if Floyd is really that bad, it's safe to say that one of those two would be the best choice for the 5th starter job. Right now in Sox camp, it's an open competition for that slot, so it's not as if there's some compelling reason to predict that the Sox would put one of the worst starters in baseball in for 21 starts. Obviously, if a couple guys break down, then Floyd, Seth Etherton, Kyle Snyder, Sidney Ponson or some other regrettably awful pitcher will be taking some starts, but that's true of just about any team in baseball.

To gnaw on this bone a bit more, it makes sense, for instance, to put the Cubs down for 28 starts from Jason Marquis with a 6.00 ERA while the putatively superior Angel Guzman toils in AAA. Unlike the White Sox, the Cubs have a huge, unwise investment in Marquis that gives evidence of an irrational understanding of the talent they have on hand. (Giving up Garcia for Floyd is evidence of a possibly misguided valuation of Floyd, but is in my eyes not compelling reason to think he'd be given a rotation job.) But without much other evidence, operating on the assumption that a team will make the worst possible choices seems pretty unfair. After all, even Baylor managed to get Bellhorn on the field for 521 PAs.

As for the mid-season trade thing, obviously I was being a bit unfair.
   31. Russ Posted: March 12, 2007 at 10:44 AM (#2310622)
You might be able to tinker, running more simulations under differing assumptions about playing time.


It would be interesting to see if the results changed by changing all the predicted below-replacement level performances to predicted replacement level, sort of a truncated mean model. That should be "easy" to do from a sim perspective, without toying with playing time issues.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Support BBTF

donate

Thanks to
Sebastian
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogThe Absurd History of “The Los Angeles Angels of Anaheim”
(104 - 4:39am, Jul 24)
Last: DavidFoss

NewsblogBaseball’s Best Teams Are Too Damn Good
(66 - 2:55am, Jul 24)
Last: Baldrick

NewsblogOT-NBA off season thread
(1342 - 2:15am, Jul 24)
Last: If on a winter's night a baserunner

NewsblogSonoma Stompers pitcher just 3rd woman to win pro game
(10 - 1:40am, Jul 24)
Last: Shock

NewsblogOTP 17 July 2017: Love is baseball, family. Keep the politics out
(1475 - 1:37am, Jul 24)
Last: Chicago Joe

Newsblog20 years ago today: Greg Maddux threw a 76-pitch complete game
(31 - 12:47am, Jul 24)
Last: jacjacatk

NewsblogHe played where? Majors' best 'irony jerseys'
(80 - 12:43am, Jul 24)
Last: Crispix reaches boiling point with lackluster play

NewsblogA sandwich is a sandwich, but OMNICHATTER is a meal! for the weekend of July 22-23, 2017
(207 - 11:58pm, Jul 23)
Last: Eric J can SABER all he wants to

NewsblogAlex Rodriguez has his very own arcade machine called the 'A-RodCade'
(16 - 11:56pm, Jul 23)
Last: Cargo Cultist

Sox TherapyMinor Moves
(30 - 9:57pm, Jul 23)
Last: John DiFool2

NewsblogIt's unacceptable — Cardinals have lost their Way under Matheny
(27 - 6:02pm, Jul 23)
Last: Sleepy's still holding up that little wild bouquet

NewsblogRays have the ultimate troll for visiting managers
(3 - 5:48pm, Jul 23)
Last: Rowland Office Supplies

NewsblogTaking Back the Ballparks - San Francisco Giants
(188 - 4:16pm, Jul 23)
Last: SBB, Live from the Alt-Center

NewsblogCBS Denver: Charlie Blackmon Kid Meets The Real Deal
(6 - 2:00pm, Jul 23)
Last: What did Billy Ripken have against ElRoy Face?

NewsblogNYP: Yoenis Cespedes: I want to finish my career with A’s
(17 - 8:43am, Jul 23)
Last: Tony S

Page rendered in 0.2898 seconds
47 querie(s) executed