THE BOOK--Playing The Percentages In Baseball

Sunday, October 25, 2009

What would happen if Girardi benches Swisher and plays Gardner in center and Melky in right…

Other than you would hear me scream even if you are 1000 miles away.

I have heard rumors that Girardi might bench Swisher and I wouldn’t be surprised. Even though it has been painful watching him hit in this post-season, Nick Swisher is a very good offensive player.  Period.  He is one now.  He was one 3 weeks ago.  And he was one 2 months ago.  And he likely is going to be one tomorrow and the next day.

I really don’t know how people justify the thought process that when he steps up to the plate on Sunday night and faces Joe Saunders or whoever, that somehow whether he has gotten 2 hits or 7 hits in his last 20 PA is magically going to affect whether he gets a hit or a walk during that PA or any future PA.  I really don’t.  Maybe it is just that my whole life I have never believed in that kind of mythology so that it is second nature to me to think rationally rather than irrationally and superstitiously.  For example, it would never even occur to me that if I played blackjack or poker for 20 minutes and was “hot as a pistol” that somehow I would expect to win more than I was supposed to on a probability basis for the next 5 or 20 minutes.  That would be magic or voodoo to me, and I don’t believe in magic or voodoo other than the David Copperfield kind.  I mean, how would that even be possible?  Somehow, inexplicably in my opinion, otherwise intelligent college graduates, lawyers, doctors and engineers believe in that kind of magic and voodoo.  I can’t for the life of me fathom how they can.  I really can’t.  What mechanism do they think is at work?

I would love to sit down with any of those folks and have them explain to me how exactly how that could work.  How could a dealer deal me a hand of blackjack after she just shuffles a deck of cards and the result of that blackjack hand have anything to do with the results of my previous 10 or 100 hands?  How could that POSSIBLY be?  I know I am preaching to the choir on this board, but I would like to ask those questions to one of the multitudes of lawyers and doctors and engineers who actually believe that to be the case, or at least speak like they do (if you actually got them to think about it for a minute, many of them would realize how silly the notion was).

Anyway, the Vegas “no-juice” line for game 6 is around NYY -173, which means that they have a 63.4% chance of winning the game.  Let’s use that as the baseline assuming that Swisher is playing.  How does that change if we insert Gardner in the lineup?  We’ll turn to my trusty sim to figure that out.

With Gardner in center and Melky in right, that number goes down to 61.83% or a loss of 1.6%. That is a pretty big difference.  That would be almost equivalent to pinch hitting for A-Rod in the 9th inning of game 5 and about the same as IBB’ing A-Rod (for the Angels) in terms of the loss in WE.  The problem with playing Gardner in game 6 is not so much that he is a bad player.  He is not a bad hitter actually and a very good fielder, especially as a RF’er.  It is that he is a lefty facing Saunders.  What about if we play Hairston rather than Swisher?  Hairston is actually a worse hitter overall than Gardner, but is a RHB of course. His fielding in RF is also excellent, at least according to my defensive projections.  Anyway, with Hairston, the WP is 62.3% or a loss of 1.1%, a little better than playing Gardner.  And BTW, the sim uses base running skill and speed in its computations.

Yankee fans better hope that this rumor is false and that Girardi doesn’t make his biggest blunder so far of the post-season.