THE BOOK--Playing The Percentages In Baseball

Since the Steinbrenner news broke, I have no doubt WFAN will cancel all non-Steinbrenner appearances.  WFAN is great at getting all the Yankee stars of the past to come on the air in times like this.

I have not turned WFAN on since Mike / MadDog broke up (I could justify listening to them because one would try to talk sense to the other, but not since).  Today will be an exception I think.

On #3, Maddux is not really a good choice to make this point.  His BABIP was .009 better than league average, and saved him about .20 R/G.  That accounts for 20% of his advantage over an average pitcher.  Not quite as important as his individual “DIPS” stats, but a significant part of his success.

Games are decided by runs, which are a statistic.  If you are anti statistics, then you should be wanting baseball games to be decided by judges on artistic merit.  If not, then you are statistically inclined.  You are merely arguing one statistic over another.

One of my talking points I make in a discussion of pitchers wins is this:

“Winning the game is 50% runs scored, 50% runs allowed.  A pitcher can only affect how many runs he gives up.  How many runs he gives up is part how well he pitches, part how well the defense fields, so ultimately, a pitcher has less control over his W-L record than the offense does, since offense is 50% and the other 50% is split between pitching and defense.”

And then something about how sabermetrics is just coming up with better ways to evaluate how an individual’s performance helps the team win - like figuring out how much a pitchers performance actually helped his team win instead of using W-L record, which again, is more influenced by the offense than the pitcher.  So that’s what metrics like FIP, xFIP, Siera, etc are doing for pitching, and what linear weights does for hitting (instead of stats like RBI’s and runs that are largely affected by teammates).

This is something I’ve wondered about. Let’s say that Greg Maddux’s career BABIP is below what we would expect, giving his park/fielders/etc. Does that tell us anything about how Maddux pitched? I know that Gold Gloves are a weak data point when it comes to fielding analysis, but the guy won 20 of them.

Nick: I don’t disagree with your position.  Insofar as what the guy is asking from us, it has to be positive.  You can’t put the radio host on the defensive.  MGL’s actually done several radio appearances, one with Richard Justice who he tangled with online pretty strongly, and Justice was very welcoming of MGL.

It’s really surprising the difference in the way people interact when it’s one-on-one, as opposed to each side getting the debate floor to themselves.

So, the key is to have a dialogue, and the other guy has to be along for the ride.

re: #4 Hot/Cold Streaks

Count me among the unwashed masses who have trouble believing that recent performance is irrelevant in predicting the coming month’s performance.

For example, should I believe that Alex Rios is no more likely to have a good next month this year than he was to have a good September last year?

I am fascinated by the subject.  I’d really appreciate it if anyone could point me to studies on the subject.

Thanks.

"For example, should I believe that Alex Rios is no more likely to have a good next month this year than he was to have a good September last year? “

No, you should not believe that.  Three months of good results will change our expectations of him.  But we shouldn’t just forget about 2009, either.

Before the seasons, ZiPS projected a .275/.325/.442 line for Rios.  With knowledge of his 2010 season so far, it now projects a .284/.338/.470 line going forward.

Hey Tom, thanks for the link. I received a call from the producer Pete Bilotti this afternoon, and it’s been rescheduled for 2:00 on Thursday (7/15). Hope your readers can tune in, but again I have my doubts about how much depth we’ll get into. I think I’ll just make the case that if fans want to criticze their GM, they should put themselves in his chair. Once there, SABR tools are a great way to asses value to a player, sort out luck from skill and find undervalued assets.

Bill/7 I’m sure someone will correct me if I’m wrong.

It’s not so much that hot and cold streaks don’t happen.  Guys might be hitting poorly because of a mechanical issue in their swing, because their muscles are sore, etc.  I don’t think anyone’s disputing that those things affect a player’s performance.  The problem is that, looking at a player’s performance, you can’t distinguish that from random variation, when you land tails 10 times in a row.  That happens, too.  And of course, players will give reasons regardless of whether there are reasons (after all, people pester them with questions).  So, looking at the data, you can’t predict a change.  That’s where a look at batted ball data and the eye of a good scout can really inform.

I guess my point is that “no predictive power” does not mean “doesn’t exist.”

P.S. They covered the topic in The Book, but I don’t know of any other studies on the web.  They probably exist.

Re: #1, RBI

Here’s a demonstration for how important teammates are to RBI production.  In the American League, a position player bats ninth.  In the National League, a vastly inferior batter, the pitcher, usually bats ninth.  Here are last year’s PA and RBI totals for AL and NL leadoff hitters:

AL:  968 RBI, 10,683 PA
NL:  938 RBI, 12,195 PA

AL leadoff hitters drove in more runs despite well over a thousand fewer trips to the plate.  You can’t bring in a runner if there is no runner.  Move that NL leadoff hitter to the middle of the order and watch his RBI totals climb.

#11: And in the NL, you’ll often have the pitcher sacrificing runners along an extra base, even.

That means we should strike Rickey Henderson and Tim Raines RBI totals from the record book because they didn’t have a level playing field.

(Applies to the Jeter Hits thread.)