2009 Bill James Handbook Projections
They’re here!
The 2009 Bill James Handbook projections are now available on FanGraphs in the individual player pages. Big thanks to Baseball Info Solutions for letting us run these for the 3rd year in a row! If you want the sortable stats version and are not content with seeing one player at a time, you will have to purchase a copy directly from them.
David Appelman is the creator of FanGraphs.
Awesome, David.
Very cool.
So is James really projecting Matt Kemp to have a .372 BABIP but Andy LaRoche’s BABIP to be only .270? Really?
I mean, I understand that Kemp has demonstrated an ability to hit line drives, but over .100 points of BABIP seems a bit extreme.
To be completely transparent here, there are no actual BABIP projections that they provide, but when you calculate BABIP off the projected stats, that’s what you get.
So essentially, it’s a projected BABIP of .372, which to me does seem a bit out there. Even someone like Ichiro can’t realistically maintain a BABIP above .350.
Overall though, this seems to just be a playing time issue. They seem to regress his At-Bats and leave many of his raw stats the same or make them better. If you adjust his playing time to be what it was last year (which to me seems somewhat likely), his rate stats will also remain the same.
David, I noticed new “Plate Discipline” stats for pitchers. I’m drooling at the prospect of even more stats to analyze, but I’m not sure exactly what each category means and didn’t see a post from you explaining the addition of these new stats. So tell us about them!! Thanks.
So then, breaking it down to see what the BAPIP figures would have to be to fit the projections sort of identifies some problems with James’ methodology, don’t you think? Not to be hard on James – I don’t think PECOTA takes BABIP into account either and we know it’s overall a very good system (right?). But it looks like an area where projections could be improved, no? Anybody know if CHONE or THT’s projections properly calculate (based on batted ball data) and then regress BABIP when making their projections?
Nick, you might be interested in an article I recently posted on THT, which get’s right to the heart of predicting BABIP using key performance indicators.
http://www.hardballtimes.com/main/article/batters-and-babip/