During the offseason I had a few words for the good people who run Baseball Prospectus’ PECOTA projection system. In short, while every other projection system predicts a first-place Yankee finish, PECOTA projects a third-place, not-in-the-playoffs finish. Also, BP has constantly updated its forecasts, apparently to rip off anyone who purchased their book — unless it’s also self-updating. You can find my thoughts here.
Well, BP is at it again. As of April 10, they’ve updated PECOTA once more. This was a welcome surprise. Before I clicked on the link I was excited to see the results, since on April 10 the Yankees were off to a 3-2 start and had handled the Red Sox at home. In response to the strong performance, Baseball Prospectus had revised their estimate of the Yankee season DOWNWARDS. Previously they’d predicted 91 wins for the Bombers; now they predict 89. The team still misses the playoffs.
My reaction is to hold BP accountable. Throughout the season this will be a running theme on the site. At the end of each full month of baseball I’ll run through my predictions and PECOTA’s predictions as of April 10 to see who is more accurate, thus far in the season (my predictions are in the link above). I’ll also throw out some of PECOTA’s better gems of wisdom.
It’s too early to compare my forecasts to PECOTA’s. We’ll do that for the first time on May 1. But, it’s not too early to point out how ridiculous it is to make predictions on April 10 that are completely devoid of reality. Without further ado, a sampling of PECOTA’s greatest hits:
Yankees | Royce Ring P | 19 innings of work. I don’t even know who this guy is.
Yankees | Mariano Rivera P | 2.56 ERA. This isn’t a bad projection, but it’s not great for Mo, who is off to a torrid start. Shouldn’t his increased velocity factor into their projection? Is anyone else angry about this?
Yankees | Javier Vazquez P | 3.75 ERA, 1.2 WHIP. Yeah, this just isn’t going to happen. What makes the projection particularly galling is that it came on the heels of Javy’s awful season debut. Both numbers are also better than what CC is projected to do. They are checking these figures, right?
Red Sox | David Ortiz DH | .264 AVG, 24 home runs. This won’t happen if he’s benched. Right now Red Sox nation would thank god if Papi could muster such a solid season.
That’s just a taste. As always, my gripe isn’t that they’re short-changing the Yankees. That’s their system and their prerogative. My gripe is that they keep updating a system that has a corresponding hardcopy version and the updates don’t seem to reflect reality at all.
PECOTA, on May 1 — it’s on!
If you enjoyed this article, please consider sharing it!
4 Responses to PECOTA and accountability
Leave a Reply
TIQ IQ
While I'm not a Baseball Prospectus subscriber and therefore didn't plunk down any of my hard earned scratch, I am a fan of statistical systems AND it's annoying to see a system that makes no sense that people are paying for and people talking trash about my bomber boys (as Joe West learned, Mariano River doesn't play that, and neither do i!)
So Baseball Prospectus updated their system on April 10. Logic dictates that there are only 4 reasons to update on this day:
1. incorporating spring training data
2. incorporating spring training data and early season data
3. incorporating early season data
4. they found a bug and they patched it
the first 3 reasons are dumb reasons to update. for 1, spring training data is pretty useless, especially for pitchers who might be trying to add a pitch and are in no danger of not making the team (AJ Burnett)… or say for Andy Pettitte, who hardly got a chance to pitch at all in the spring. if they are doing reason 2, i would say those are both 2 very small and some what useless sample sizes they added to their data. if reason 3, then i'd say they're incorporating a sample size that is WAY TO SMALL. and reason 4, well, this sample data you provided tell me their system is still buggy as hell. i've seen Royce Ring's name in print, but never seen him pitch. I'd say he's a Sept call up at most, unless disaster strikes…
BP is bizarre. i dont understand what they're thinking… although that number on Javy might be fair… he had a 2.25 last year, right? i think i read somewhere that the standard increase from the NL to the AL is .75, add another .25 for being the AL East and add another .50 for him pitching above his ability last year and against inferior competition…
but yeah, i dont understand how BP's system works. i think if they could be a bit more forthcoming, that would elevate some frustration
~jamie
BP, are you with me?
But seriously, it's always fun to come check out this site and find a post from Michael Jaggers-Radolf about his problems with BP. For further reading please look here, here, here, here, and here.
I'd like to clarify that I'm not a supporter of BP and their calculation methods; I'm a hater of MJR and his tiny penis. Watch Justified on FX!
You keep making straight to DVD classics like Hit Man and growing your career to new heights on the bowels of basic cable and I'll continue to write about whatever I want.
PECOTA can suck it. I know we've discussed this ad naseum on the yanks list but giving them any credibility is like saying Spring Training stats mean something.
Take a look at the ESPN Standings widget on the right-side of the page. The Yankees are tops in the division, and they will be at season's end.