Proceeds from the ads below will be donated to the
Bob Wuesthoff scholarship fund.
Friday, April 03, 2009 |
Angels Suck Even More Than You Thought Possible, Says PECOTA
81-81 record, second place to ... the A's.
Yup, the season doesn't start any different this year, either.
The Dodgers are projected to 92 wins, 4 better than the second-place Diamondbacks.
Labels: angels, dodgers, predictions
Comments:
I call shenanigans. This team outperforms their pythag every year. 86 wins will win the division by 4.5.
Feh on PECOTA. Didn't they have the Angels at 79 wins last year? Didn't they beat that by, oh, 21 games? Figure between 91 and 94 wins, then, assuming that two of the three starters are back by June.
Which means, if PECOTA does as well as it did last year, the Angels will be a 96 run team.
If the system keeps failing, then the system is no good. They had the Angels at 85 wins last year...a full 15 games off the mark. The average error for team W-L predictions in 2008 was 8.5 wins. That's pee your pants in public bad. From 2003 through 2008, its average miss was 6 wins. Meaning, peering through PECOTA's margin of error, one can hardly tell the difference between a 95 game winner and an 83 game winner.
Now, I think PECOTA is a relatively good prediction system for individual players who have had 3-5 years of recorded MLB playing time. But, because it's bad in predicting rookie performance, cannot anticipate injury, and does not accurately factor in the likelihood of young players ascending to the club mid-season and making an impact, it makes all sorts of bad guesses in aggregate.
If the system keeps failing, then the system is no good. They had the Angels at 85 wins last year...a full 15 games off the mark. The average error for team W-L predictions in 2008 was 8.5 wins. That's pee your pants in public bad. From 2003 through 2008, its average miss was 6 wins. Meaning, peering through PECOTA's margin of error, one can hardly tell the difference between a 95 game winner and an 83 game winner.
Now, I think PECOTA is a relatively good prediction system for individual players who have had 3-5 years of recorded MLB playing time. But, because it's bad in predicting rookie performance, cannot anticipate injury, and does not accurately factor in the likelihood of young players ascending to the club mid-season and making an impact, it makes all sorts of bad guesses in aggregate.
love the blog. go dodgers. i just did a preview of the dodgers on my blog if you wanna check it out! www.baseballtalktv.com peace!
# posted by 
: 4/04/2009 02:22:00 PM
: 4/04/2009 02:22:00 PMNewer› ‹Older
 |
Post a Comment
Note: Only a member of this blog may post a comment.