Fun With PECOTA (Part 2)
Last week, I had a little fun with Baseball Prospectus' 2012 PECOTA foreshadowing system. I'm calling it a foreshadowing system because the word forecasting implies something a little more definite. After all, meteorologists are forecasters, and since they are infallibly nostradomic, it would be improper to apply the word to a subject as nebulous and chaotic as professional baseball.
In all seriousness (or at least more seriousness), I would like to say that I am in awe of the amount of labor and thought that goes into the creation of PECOTA every season. It can't be easy to build. With thousands of players to account for, with each one's likelihood to do, well, whatever it is that they might do, there is a staggering amount of data to process, and to be snide without at least tipping one's hat would be unfair.
PECOTA does one thing extremely well: It gives people a broad picture about the abilities of a player to impact his team. It is a guide through the future season; a guide who knows vaguely what direction we are heading, but not the names of the roads to be taken.
But PECOTA needs numbers to present that overview, and the numbers are where the fun is. So after the jump, I will continue my good-spirited and well-meaning mockery by examining how PECOTA views the Astros' pitching staff and its overall win-loss record.
Note: I want to point out the totally awesome photo of Wesley Wright. I don't know who Julio Cortez is, but that's a really fine example of photographication, and he deserves kudos for taking this one.
Like any good engineer, I prefer to present my data in the form of a bullet list. So here is the Astros' pitching staff, according to PECOTA.
- The lowest ERA on the team will belong to Brandon Lyon (4.01) Even more shocking, PECOTA thinks Lyon will manage a staggering 38 innings pitched. That's like 39 more innings than last season.
- Wandy is the only pitcher (besides Lyon) with a FRA (Fair Runs Allowed) lower than 5.00. FRA is like FIP (Fielding Independent Pitching). Don't ask me to explain the difference, but it uses the same scale as ERA, so it suffices to say anything over 5.00 is bad.
- PECOTA projects 1,444 and 1/3 innings pitched by Astros pitchers this season. According to my handy calculator, that equals 160.5 games. That means the Astros will forfeit 1 game and have another rained out after 5 innings. And, there will be no extra innings games.
- Most likely to break out: Jordan Lyles (32%), Fernando Rodriguez (30%), Bud Norris (29%)
- Most likely to beat PECOTA projection: Bud Norris (67%, 4.77 ERA), Jordan Lyles (66%, 5.29 ERA), Zack Duke (61%, 5.53 ERA), Fernando Rodriguez (59%, 5.59 ERA).
- Most likely to collapse and fall below PECOTA projection: Brandon Lyon (41%, 4.01 ERA), Sergio Escalona (36%, 5.04 ERA), Juan Abreu (34%, 4.98 ERA)
With any luck, PECOTA will be no more accurate than Chuck the Weatherman, and the Astros will have sunshine and blue skies (metaphorically speaking) for all of 2012.
25 comments
|
0 recs |
Do you like this story?
Comments
A 102 loss bottom line isn’t unrealistic. But some of those pitching projections strike me as seriously pessimistic. Sounds like PECOTA doesn’t like any of the Astros’ pitchers (other than Brandon Lyon!).
PECOTA and Wandy
It looks like PECOTA is tough on pitchers in general this season. Something in the algorithm. Wandy’s ERA is projected 4.33 (Which I think he’ll beat), but he’s given a 46 “Improve” score, which is pretty high. PECOTA is saying Wandy has a 46% chance to improve on his PECOTA 50th-percentile projection. For a pitcher in his 30’s, a 46% chance is very, very high. For comparison, Tim Stauffer (4.16 projected ERA) only has an 11% chance.
In otherwords, PECOTA is recognizing that its relatively high ERA for Wandy is affected by external things such as past injuries. It can’t mathematically take such abstract things into consideration, but it does seem to recognize that those things are there and provides a likelihood that an absence of those things will cause an improvement over his projected stats.
Looking deeper
The pitchers I’m seeing who have the highest IMPROVE scores are the studliest of studs who routinely beat their projections and guys who have missed a good chunk of time with injuries over the past two seasons.
The studs get a good IMPROVE score because PECOTA weights towards the average, which brings down projections for superstars. It’s recognizing that these guys have the talent to blow away a projection that is brought down mathematically.
The injury-past guys are getting a high IMPROVE score because PECOTA recognizes that their stats are calculated using essentially incomplete seasons’ worth of statistics.
For what it’s worth, Bud Norris’ projected ERA is 4.77, but his IMPROVE score is a whopping 68%, which is the 3rd-highest out of pitchers projected to pitch 150+ innings. That’s a good thing.
How can it make all those comparisons, yet still project virtually every one of our pitchers to such high ERAs?
For the starters, I think Wandy and Norris will both be in the 3.50-4.10 range, with Myers and Happ in the 4.10-4.80 range. The 5th starter should be in the 4.50-5.30 range.
Escalona, Lopez, and Carpenter will all have ERAs under 4. Whoever makes up the rest of the bullpen will be wildcards but I doubt more than 2 or 3 will have ERAs over 5.
Wandy’s good years may be behind him so it’s possible he could become a 4 ERA pitcher.
Follow my ramblings on Twitter .
by Timothy De Block on Feb 23, 2012 12:48 PM CST up reply actions
What’s funny is that in real life I’m a horrible pessimist. Only my internet personality can see the sunshine.
by Snake Diggity on Feb 23, 2012 2:24 PM CST up reply actions
looks like roy is the new roger
by conroestro on Feb 23, 2012 2:25 PM CST via mobile reply actions
i thought the point of signing a one year deal
Was to prove he can stay healthy playing a full year.
by lawson3 on Feb 23, 2012 3:06 PM CST via mobile up reply actions
he’s gotten offers just none that he likes.
Follow my ramblings on Twitter .
by Timothy De Block on Feb 23, 2012 4:12 PM CST up reply actions
i guess he hasnt recieved any
Tractor or Bulldozer related incentives.
by lawson3 on Feb 23, 2012 5:09 PM CST via mobile up reply actions 1 recs
Brewers get Braun back
by conroestro on Feb 23, 2012 4:24 PM CST via mobile reply actions
Ha. I apparently did not dig into this story, enough. Well hearing that it was a decision made by independent arbitration, I guess I’ll trust that the fix wasn’t in. I had initially thought MLB made the decision.
Lyon
This article by Jerome Solomon on Brandon Lyon’s attempt to recover from arm surgery is interesting. Lyon talks about pitching through pain the last two years with the Astros. Remember when he had the shoulder cyst removed at the beginning of his first Astros’ spring training? Doctors told him he had a torn rotator cuff then. He found ways to pitch through the pain in 2010, and thought he could do the same last year. He kept changing his arm slot to reduce the pain, but he couldn’t get his pitches to work anymore. Now Lyon says he is completely pain free for the first time in many years. But he has to work on eliminating all of the bad habits (like different arm slots) that he developed with the injury.
I think Brocail may be a good pitching coach to help a rehabbing pitcher like Lyon. Brocail overcame more arm surgeries than most any pitcher I can recall. Brocail says the rehab time for shoulder surgery is a year-plus and for Lyons’ surgery it might be a year-plus-plus. He is trying to keep Lyon from doing too much too fast.

by 

























