Bill James Projections
by davidsabin on Nov 18, 2008 11:08 PM EST
Email
PrintI have been looking at some of the Bill James projections for 2009; I have never really used them before. It seems like they are a little more on the liberal side as compared to Marcel and Chone. I think I would give him credit for at least trying to incorporate some common sense rather than just spitting out numbers.
What are your opinions on the Bill James projections? Some projections will be off, I realize this, but from past years does he seem excessively optimistic or in the ball park?
4 comments
|
0 recs |
Do you like this story?
Comments
Excessively optimistic
If every hitter performed at Bill James’ level, you would see almost every team in baseball score 1000+ runs. Take his predictions with a grain of salt.
I find Zips to have a lot more value, and those stats are free. If you want to pay, PECOTA is the best around IMO. And you cannot beat the player cards, with trend analysis, likelihood of breakout/decline, etc.
Bill James Projections
Three things to note:
1) Theyre usually excessively optimistic, but its a good way to find sleepers.
2) Rookies typically get eye-poppingly good projections. Take them with a grain of salt.
3) Ignore the pitcher projections – James doesnt have much to do with them. He explained before that he doesnt believe it was productive to make pitcher projections because theyre so inaccurate, but he lets other people come up with a set (usually similar to ZiPS & CHONE) because people expect them. The hitter projections are his bread and butter.
4) One strength is the projections for stars/superstars. I tend to find James does best on these. ZiPS & CHONE both seem to lose perspective after a point and give in to a steep regression to the mean. Thus ZiPS & CHONE never seem to predict the league leader to have more than 32 HRs or so… or 105 Rs/RBIs.
The James projections have their weaknesses, but if you understand them they can be really useful. I do look forward to them coming out each year. I do think ZiPS is the best of the free systems. I subscribe to BP and get the PECOTAS (BP is well worth the money!) and those are much better than the other projections, but they dont really have a single number set/slash stat line. Everything is expressed in terms of probabilities. They do put up a weighted mean projection line for fantasy purposes… and that’s the real prize. The PECOTA weighted means trump all my other data evaluating players.
Well, you also have to realize
That marcel and Chone and pecota are pessimistic because they regress everyone to the middle, so basically they are designed to be statisticall accurate… e.g. they will nail the middle 68% to a T, but will struggle on the high end and low end of the spectrum.
The truth is that all the projection systems end up really really close in over accuracy, no matter how you measure it… The beauty of HQ, is it moves you away from projections to buying skills…. Because projections are inherently inaccurate, pay for enough skills and you will get what your looking for.
Check out my baseball analysis blog FANalytics
BaseballHQ
What kind of stuff is available this time of year? Im strongly considering subscribing – based on the reviews you guys just gave me – but should I wait until like February or so? Is there new content year round…? Or do they start to gear up for 09 at some point…?
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
