On 3rd Base Coaching Decisions
by, 07-22-2013 at 11:55 PM (1099 Views)
Letís start with this: I donít care if Aaron Hicks slid or not Ė Joe Vavra blew the call to send him last night in the fourth inning. This is not debatable. I doubt Vavra would debate it. And even had Hicks slipped into some shiny leather and slid like Kate Beckinsale in Underworld, he was still going to be out. It was not close. It was Vavra's mistake that the announcers should have been talking about.
Making outs is a cardinal sin for a third base coach. (Technically, I forget whether itís a subset of ďgluttonyĒ or ďgreedĒ.) But in this case, this story is not to beat up Vavra about a bad call. Itís just to explain the philosophy a third base coach must have, and the math behind why he must have it. And maybe learn why Vavra could risk a little more on this play.
The basic rule for a 3rd base coach is if you think thereís a decent chance your runner can get thrown out, stop him. This can be shown mathematically, which Iíll demonstrate below, but it also just makes sense: the value of the extra base, even though itís tied to a run, is nowhere near the cost of adding an out AND losing a baserunner at 3rd base.
Sabrmetrics has provided a lot more precision, and it lends a little extra insight in this case. In this game, because the hitter can advance into scoring position on a throw to the plate, that rule is not quite as stringent.
Sabrmetrics, for problems like this, uses something called Palmer and Thornís Run Expectancy Matrix. Pete Palmer and Roger Thorn studied 75 years worth of baseball games and found out the average number of runs that scored in basic situations. (By the way, they did this back in 1975.) You can find it here, but these are the prevalent numbers:
- The average team with runners on 1st and 3rd and one out will score 1.088 runs in that inning. Thatís what happens if Vavra holds up the ďstopĒ sign.
- The average team with a runner on 2nd and two outs will score .348 runs in that inning. Thatís what happens if Vavra waves Hicks home and heís thrown out.
- The average team, with a runner on 2nd and one out will score .699 runs in that inning, PLUS they would have already scored a run, so thatís 1.699 runs. Thatís what happens if Vavra waves Hicks home and heís safe.
So if Hicks is sent home and safe, the Twins gain about .6 runs. If heís out, they lose .75 runs. It doesnít take a math major (just a good algebra background) to see that Hicks needs to be safe about 55% of the time to break even. If Vavra felt it was a ďcoin flipĒ situation, sending Hicks is defendable. (It didnít look like it was, but given Hicks speed, maybe he had additional confidence.)
Now thatís just the base rule. It assumes that there are average hitters and average pitchers and average fielders, etc. In this case there were some extenuating circumstances.
For starters, the next hitter was not average. In face, Doug Bernier didnít have a major league hit and he's 32 years old. He might be more likely to strike out or to hit into a double play than an average hitter, so maybe sending the runner makes more sense. Of course, Joe Blanton is on the mound, and he hasn't been an average pitcher. He might be more likely to give up a couple more hits, so maybe it's a better idea not to send the runner. Finally, batting behind Bernier is Joe Mauer, who is a pretty good guy to have up in a clutch situation, which is probably the best reason to keep everyone from risking that extra out. So in this game, itís hard to find extra incentive to risk that out.
But the decision to send Hicks by Vavra might not have been as egregious as it initially looked. (And certainly not as bad as it looked after Bernierís double.) Third base coaches need to be pretty conservative in general, but as far as picking a moment to be aggressive, this was a pretty good choice.