Monday, April 02, 2007

Waving In Eckstein

With one out and down four runs in the sixth inning in Sunday night's opener, Cardinals third base coach Jose Oquendo sent David Eckstein home from second on a single up the middle. The results were predictable. No, I didn't have the clairvoyance to predict Carlos Beltran's awesome laser to the plate to nail Eckstein, but I did lay 100-1 that ESPN would praise Oquendo's aggressiveness if Eckstein scored while criticizing it if he was thrown out. As with Mike McDermott at the end of "Rounders", you knew who was going to walk away with all the chips, but it was fun to watch anyway.

So, was Oquendo's decision to channel Wavin' Wendell Kim the right move? Probably not. This is the kind of analysis that Win Expectancy was designed for. Courtesy of WalkOffBalk.com:

Cards' W% if:

Eckstein held at third: 16.9%
Eckstein scores, Wilson to second: 17.3%
Eckstein thrown out: 7.7%

Now, these figures have to be taken with a grain of salt, because there is an inadequate sample size, particularly in the first case, to determine the actual probability of winning with great precision. Nevertheless, it is striking how little Eckstein's run actually meant. This is because the Cardinals need a big inning to win the game, and their chances of getting a big inning are virtually the same whether Eckstein scores or not. His out, however, largely killed their chances of a rally.

I'm not a third base coach, but I probably know more about optimal math-based baseball strategies than Jose Oquendo, and I'll bet if he knew Eckstein would have to score more than 90% of the time to make this a good play, he wouldn't have looked like this. Oquendo can't be expected to generate this exact figure in his head, but someone in St. Louis should teach him some basic probabilities to use when he's on the field, and combine them with his best judgment.

On a side note, I agree 100% with Beltran's decision to throw home. The Cardinals' win probability holding Wilson at first is 17.1%, a cost of just .2% compared to a potential gain of nearly 10%. This squares with logic; Wilson's advance to second is no big deal in a 3-run game, but throwing Eckstein takes away both a run and one of St. Louis' precious remaining outs. The only other play Beltran should even consider is to fake like he'll airmail the cutoff man, hoping to catch Wilson in a rundown.

No comments: