A bunting situation

[brett]

Today the Rockies had runners on first and second with no outs and red-hot Seth Smith batting. They were down by 2 in the 7th at the time (I think).

They chose to have Smith bunt the two runners over.

If Smith is successful, is this a good move in general? Is it a good call (considering the chance for it not being successful?) Is it good pretty much in general, or only because CO was down by 2 or close at the time?

Discuss: Is bunting with no outs and runners on first and second a good move and if so, when and by how much in terms of run expectancy.

Jim Tracy by the way really seems to understand percentages in those situations, and he mentioned the prevention of the double play possibility and other factors. Do baseball managers have books that give them quick percentages for run expectancies and things?

[538280]

According to the numbers given on Tango's website from the years 1999-2002 (you can see that here) teams with runners on first and second with no outs subsequently averaged 1.573 runs in the inning, while teams with runners on 2nd and 3rd with no outs averaged 1.467. So in terms of which is going to net you the most runs overall that data suggests it's not a great move, though of course there are other factors such as who is up and who is on deck, etc. There also is another factor which is the inning and the score, there are some times where playing for just one run may be worth it rather than just going for the most runs overall. The book Baseball Between the Numbers has data for the 2005 season on the percentage of times a team scored at least one run in each base/out situation. That shows that with runners on 1st and 2nd, no outs teams scored 61.6% of the time while with runners on 2nd and 3rd they scored 67.4% of the time. So if you need only one run it probably is a good strategy, though the fact that it might not be sucessful may cut into the percentages for it being a good move. Being down by 2 and with still two innings after that I don't know if it was the greatest move though of course there are many factors outside of just these numbers specific to that and every game.

[AstrosFan]

On the run frequency tables, however, the chance of scoring two runs in a first and second, no out situation is .165. In a second and third, one out situation, it goes up to .218. Change two runs to three runs, and it goes to .127 and .101. So it looks like you'd have a better chance of tying the game in that inning by making that move, but a worse chance of taking the lead.

[Ubiquitous]

frequency tables, linear weights, all of it are useless when talking about specific situations in a specific place with specific people at a specific time.

[mikefast]

You also need to account for the probability that the defense will not record an out on the bunt attempt. It happens often enough that it significantly changes the calculus.

The Book (by Tango, Lichtman, Dolphin) covers the strategies of bunting in excellent detail.

[brett]

The Rockies were down by 2 and in a 3-1 game. Both baserunners were fast, and the 2 run probability probably depends a lot on the runner moved from first to second being fast enough to score on most singles.

It probably also depends on the groundball tendancies of the pitcher.