The thought process behind the Bill James system

image
Are the Bill James projections for a guy like Matt Holliday directly comparable to other systems after all? (Icon/SMI)

A couple weeks ago, I penned an article titled “The ‘optimistic’ Bill James projections.” ‘Optimistic’ is in quotation marks because I posited that the James projections may not be any more optimistic than CHONE or Oliver or any other system, but rather they just seem to be as a result of context. Once we compare the James projection for one player to all of the other players James projects, we will still come up with relative rankings that should be on par with the other systems.

As I’m in no way affiliated with James himself, the projections, or Baseball Info Solutions, there were some things I wasn’t entirely clear on and some questions I couldn’t answer. Luckily, a member of the BIS team who works on the projections (Ben Jedlovec) read the article and was kind enough to offer an explanation for some of the things I and the THTF readers noted. I’ll post his two e-mails in full.

Derek,
I saw your article today on the Bill James projections and wanted to say that I enjoyed your perspective. As someone who had a hand in this year’s BJHB projections, I wanted to add a couple more thoughts for your consideration.

First off, we are admittedly optimistic with playing time projections, especially in the Handbook edition of the projections. The thought is to give the reader an idea of what we think a player can do if given a full season of plate appearances. We could project Jason Heyward for 200 ABs with a couple homers, but that doesn’t tell you much about what we think of his ability. We want to give him a full season of ABs to show you what we think he can do if he gets a chance.

Also, it may seem that we have a high run environment. Part of this is due to the fact that we don’t project every marginal major league bench player who will accumulate 100 or 200 ABs over the course of the season. Chances are, some of the guys we project for 500 ABs will get hurt, and 300 ABs will go to a more marginal player or a reserve who we don’t have projected. Of course this will decrease the run environment.

Keep up the good work at THT,

Ben Jedlovec
Research Analyst
Baseball Info Solutions

And here’s the second one:

Derek,

After reviewing the comments on your original article, I wanted to mention one more thing in case you do a follow-up piece: Bill is most certainly still involved in the projections. He and John Dewan worked on the original coding shortly after BIS began publishing the Handbook, so most of the work is just the mechanics of running the code. Any changes to the code are instituted by Bill and John, and Bill, John, Steve Moyer, and others review the projections at least twice each and make manual playing time adjustments before we run the final version for the Handbook.

Beyond that, I’d just reiterate what I said earlier.

Take 10 elite hitters with a similar health record- say nine of them will get 600 ABs and the other (we don’t know which one ahead of time) will get hurt in the spring and get zero ABs. We could regress them all and give all 10 550 ABs, meaning we’re selling nine guys a little short and missing big on the 10th. Or, we can project all 10 for 600 ABs, sell nobody short and missing big on the same guy who was injured.

The second approach will get you a higher run environment, of course.

For fantasy baseball, you could make a strong argument for the second approach. Both methods will miss on the one injured player, but the first approach sells nine other guys short. If both systems are roughly equal assessments of player ability (prorated to a certain number of ABs), why would you use the first one?

ON AVERAGE, the first approach will be spot on. However, the second approach will be better nine times out of 10. This is another case when the average (expected value) doesn’t give the full picture. It depends on what you’re looking for, I guess.

Ben

So there you have it. Some insight into the thought process behind the Bill James system from someone working on the inside.

Given what Ben says about the inner-workings of the Bill James system, it appears that the projections for most veteran players should indeed be comparable between projection systems (at least as far as any projection system goes — they all assume a slightly different league average from each other). The James system might only seem more optimistic because: 1) they actually are optimistic about some rookies, 2) the run environment is inflated because a lot of veteran bench-type players don’t get projected, and 3) the run environment is further inflated (over previous seasons) because the system doesn’t attempt to forecast injuries, opting to give a relatively higher number of players a full season of at-bats.

Whether or not this means we actually can compare, say, Alex Rodriguez‘s Bill James projection to his CHONE projections isn’t 100 percent certain — we’d need to run some tests, as we would with any other fantasy baseball theory — but I think Ben provided us with some very interesting food for thought.

As always, if you have any questions or comments, feel free to let me know.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Cannons in the bushes: rating minor league outfielders’ arms
Next: TUCK! sez: Don’t stop ‘til you get enough »

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *