Proceeds from the ads below will be donated to the Bob Wuesthoff scholarship fund.

Wednesday, January 26, 2005

Self-Embarrassment Time: PECOTA 2005 Projections

As you probably know by now, Baseball Prospectus has their 2005 PECOTA projections out. Hanging my fanny wide, wide, wide on the wire, its projections for the AL and NL West.

             O F F E N S E
       AL                    NL
Team      VORP     Team            VORP    
==============     ====================
Oakland  399.6     San Diego      302.2
Anaheim  306.6     San Francisco  237.4
Texas    277.3     Colorado       236.3
Seattle  239.7     Los Angeles    224.5
                   Arizona        168.2

           P I T C H I N G
       AL                    NL
Team      VORP     Team            VORP    
==============     ====================
Oakland  345.1     San Francisco  268.6
Anaheim  289.0     San Diego      235.4
Texas    271.3     Los Angeles    235.3
Seattle  214.7     Arizona        210.4
                   Colorado       178.8
Adding pitching VORP to offensive VORP, this leads to projected division winners as follows:
       AL                    NL
Team       VORP     Team            VORP    
===============     ====================
Oakland   744.7     San Diego      537.6
Anaheim   595.6     Sam Francisco  506.0
Texas     548.6     Los Angeles    459.8
Seattle   454.4     Colorado       415.1
                    Arizona        378.6
Some thoughts: If I'm being a total idiot by adding pitching VORP and offensive VORP, please let me know.

PECOTA seems useful on a player-by-player basis, but projecting the future of a whole division? I dunno.
One of the problems I forsee is that PECOTA seems to assume a player will receive playing time commeasurate with his abilities. Take Neifi Perez (please). PECOTA forecasts him to collect -6.2 VORP in 273 AB. But he hasn't had that few since '97; last year, he had 381 AB. What to make of that when looking into the future?
PECOTA can't make up for Dusty Baker or other acts of God. Your Mileage May Vary.
Yeah thats not a really good way to compare. VORP is very much dependant on playing time. The PECOTA spreadsheet has 35 Oakland hitters projected while the Angels only have 28. The problem with adding up VORPs from this can be seen in the fact that Oakland has 3 Shortstops projected (Crosby, Quintanilla and Rouse) for a total VORP of 55.6. A pretty nice total till you realize it took over 1000 PA's to get that high.

I think you'll have to figure out how much playing time each player is going to get and then multiply that by VORP/PA to get anything useful. For instance Dallas McPherson is projected as .272/.351/.496 but only 324 AB. If he's putting up numbers that good and is healthy he'll easily get 500+ AB.
What were the PECOTA projections for last year and how did they compare to what actually happened?
Yes, as I mentioned in my extreme naïveté alert, this is a situation where I will certainly have to whittle the list some to adjust for actual players. This was a quick-and-dirty attempt to come to an answer, but clearly more work remains to be done. For the NL West, I'll probably end up using Jon's rosters, but I'll have to invent something for the AL West.
These team figures strike me as ridiculous. But a question, is defense figured in here anywhere? I thought winshares factored this in buy maybe Pecota's different. If defense is not figured in here then I might buy this. Regardless, SF will pull a Mariner like collapse and the D's will edge out the Pods.
While this entire exercise is pointless, let’s not ignore PETCOTA's notorious trouble with projecting players returning from injury, which is the case with half of Arizona’s roster.
Christopher -- I have no idea if defense is factored in.

Richard -- Glaus: 345 AB, .281/.391/.579. I think it does, but it tends to be super-pessimistic. So Arizona probably has some upside there.

More later.

Post a Comment

Newer›  ‹Older
This page is powered by Blogger. Isn't yours?

WWW 6-4-2