Tuesday, April 12, 2011

On straws, camels, and assigning blame to a loss

Last week, after a 1-0 loss to the Indians,  I heard a few fans complaining that we lost the game because of Darnell McDonald's baserunning blunder.   Others blamed the umpire for calling him out when he was "clearly" safe.  My friend Reg, over at his Rulebook Guru's Blog has more to say about why blaming an umpire for a loss is never very astute.  But I want to touch on why blaming one player, or one play is also a sign that you probably should be watching hockey instead, because baseball is obviously way too complicated for you.


In baseball each team is given 27 offensive outs.  Without using pinch hitters, that means that each batter will have at least 3 plate appearances, usually 4 or 5, sometimes more.  If a batter goes 4 for 4 with 3 home runs and a double, and his team loses 5-4, although he probably helped his team more than any of his teammates, he is still partially to blame for the loss.  Had he hit another home run instead of just a double, the game would be going into extra innings.  I doubt anyone would think of blaming him for the loss, but he could have helped change the outcome more.   I have written before (sorry - no linky available anymore - it was a good one, I'm going to have to rewrite it and repost it sometime), on why Bill Buckner's error did NOT cost the Sox the 1986 World Series.  (And it has nothing to do with Calvin Schiraldi or Bob Stanley pitching, or Rich Gedman's unscored passed ball, but in a nutshell we had about 4,000 opportunities to win the game BEFORE going into extra innings.  There were many individual plays that changed the probable winner of the game MORE than the final play).  Looking at this recent 1-0 loss, there are many things to blame prior to Mr. McDonald:

First of all, how about EVERY single Boston batter?  We only got 4 hits, which meant there were a lot of 0-fers.  If any of those had been a 1-fer with a HR, this game might still be going on.   The pitching was pretty good, only giving up 1 run but it could have not given it up.  The run scored with the help of a stolen base (could it have been prevented?  do we blame Bard's slow delivery, or Salty's late throw?), and 2 bunts (better infield defense? or positioning?).  Ok, these are pretty general things, but a fan could say that we lost the game because our leadoff batter in the 1st inning grounded out instead of hitting a home run and that would make about as much sense as blaming it on the last play of the game.

Here is some very quick detailed analysis comparing a couple of specific plays:
Top of 8: Runners on 1st and 2nd with 1 out in a scoreless tie:  At this point, the standard probability of the Sox winning the game was 62%  (Using standard MLB Win probability tables from The Book: Playing the Percentages in Baseball).  The fact that the heart of the order was coming up improves that even more. So what did the Sox do?  Back to back soft ground outs by Crawford and Pedroia. Failing to score that inning knocked our chances of winning back down to 39%.   That's a decrease of 23%.

At the end of the game, if McDonald stays at 2nd base, we have runners on 1st and 2nd with 2 outs down by a run.   The probability of us winning the game at that point is only 14% (probably less because we have our #8 and 9 hitters coming up).  Getting thrown out to end the game obviously takes that down to 0.  That "blunder" was less costly then not knocking in the run in the 8th inning.  But nobody is talking about that squander.

And don't get me wrong - they shouldn't - that's my whole point!  Each at bat or play adds or subtracts a little bit from the complete picture.  No one play can ever make a difference.  Again, if you think a walkoff grand slam in the bottom of the 9th is the difference in a game, you're not thinking about the big picture.  You need to keep in mind the sequence of events that led to that situation.

The last straw that broke the camel's back, didn't weigh any more than any of the other ones.

No comments:

Post a Comment