THE BOOK--Playing The Percentages In Baseball

Monday, January 24, 2011

Talent or Timing?  Via Cain v Shields

From 2007-2010, Matt Cain has faced 3547 batters.  James Shields in that same time period has faced 3576 batters. Shields has started 130 games, and Cain has started 132 games. Neither has ever relieved. Per start, that’s 27.5 batters for Shields and 26.9 for Cain (and one would think the DH/PH issue likely would account for the difference).

Shields has struck out 698 batters, while Cain has struck out 697 (though given that Cain gets to strike out pitchers, it’s a bit better for Shields). There, the similarity ends.  Shields has given up 115 HR to Cain’s 77, or 38 more HR.  Excluding IBB, and including HBP, Shields has given up only 203 free passes to Cain’s 301 free passes, or 98 fewer walks for Shields. Overall, it’s a slight advantage for Cain if we go by FIP.  That is, the extra HR by Shields are largely, but not totally, offset by the much fewer walks.  (About 20 runs worse for Shields.)

However, Cain has given up 97 fewer runs (earned and unearned) than Shields, or about one run per 9 inning game!  Given such two identical composites, how does that happen?

There are three components to pitching:
1. The results you get when you don’t involve your fielders
2. The results you get when you do involve your fielders
3. The sequencing of outcomes

In the first case, the results you get are mostly tied to your own level of talent.  Basically, the outcomes of SO are almost all talent.  The outcomes of walks are mostly talent.  The outcomes of HR are alot of talent.  Adding it up, and, Cain and Shields are fairly equals.

In the second case, Shields’ has a batting average on balls in play of .305 to Cain’s .272.  That is 113 difference in hits (excluding HR), or outs, if you prefer.  But, how much of this is talent, and how much is it that Cain ended up with better results by better fielding and better timing?

In the third case, Cain’s results with men on base is much better than Shields’ results.  Again, how much is it that Shields lost it with men on base, and how much is it that it was bad timing?  Is it really a matter of talent, or timing, if one pitcher goes walk, double, out, out, out, and the other goes double, walk, out, out, out?

This is why god invented the bet.  If you really believed that there is in essence no difference in overall talent between Shields and Cain, then you should be taking bets to that effect, that Cain and Shields are expected to be on the mound a similar number of times when runs are scored.  But, if you believe that Cain is in fact more talented in getting positive outcomes when the ball is in play, and/or Cain is in fact more talented in stringing events to minimize their impact to run scoring, then you should be taking bets to that effect.

The problem is that a one-pair bet won’t prove much, because, well, anything can happen beyond letting talent drive the bet.  And waiting four years to see the results of this bet would be better, but then other things could affect things, such as a pitcher’s change in talent level.

Therefore, in order to have a good bet, we need to find 10-20 good starting pitchers who were like Cain, and 10-20 good starting pitchers who were like Shields, and then make your bet for the upcoming season.  I’ll come up with a list before Opening Day, and let’s see what you guys think.

When looking at things in context, what drives the results: talent or timing?