Dunno if this belongs here or in the Sabrmetrics section, but I place it here since it is a question of fact and not methodology.

The question has two parts:

1) Bunting to advance the runner to second:  is this, statistically, a cost-effective usage of an out?  Just because it is commonly done, doesn’t make it right.  I suppose it depends on who you have up there, of course, but taken all-in-all, isn’t it better to go ahead and swing away?

2) Bunting to score the runner from third:  what is the actual success rate of this play?  Is it so
unlikely to work that it is only an act of desperation?  You don’t see it nearly as often as bunting the man over to second.

Note: I’m looking for a statistical answer to these questions, not an unsupported opinion.

—Mal

Tango Tiger’s Run Expectancy Matrix shows that in every situation, you have a lower chance of scoring runs, and score fewer runs, after you do a sac bunt, except for scoring a runner from third, in which case you’re guaranteed a run, but hurt your chances of scoring additional runs.

Thus we conclude that the only good situation for saccing players over is if you want to try to score a runner from third in order to tie or take a lead late in the game.

It’s a little more complicated than just a simple lose-an-out, gain-a-base tradeoff, because other things can happen on a ascrifice attempt, and it does matter quite a bit who is batting.  “The Book” has an excellent entry about it.  But at the end of the day, sacrificing any remotely competant batter is usually a bad idea. 

On the converse, if you are on defense and the opposing team sacrifices, most of the time you should gladly accept an out for a base.

So has anybody computed the % success rate of attempted “suicide squeeze” bunts?  Even in todays run-heavy game, one would (intuitively) think trading an out for a run would be cost-effective, if it had any decent chance of success.

—Mal

malthaussen - 10 September 2006 06:35 AM

So has anybody computed the % success rate of attempted “suicide squeeze” bunts?  Even in todays run-heavy game, one would (intuitively) think trading an out for a run would be cost-effective, if it had any decent chance of success.

—Mal

I’m fairly that sure that the chance of success is extremely low, and at that it would be better to just let the hitter swing away. A hit or a sac fly scores the run.

I would also like to mention that the point of this discussion is to get player A to second base and another way to do it is steal it lookig at Tango Tiger?s Run Expectancy Matrix compare the runs expected with one out and a runner on second compared to runner on second no outs.  Given stolen bases may not be as successful as often as sacrifice attempts. I would like to see those numbers.
            outs

Runner 0     1       2  
Empty   X   .297   X     result if runner is thrown out
1st     .953   X       X   result if hitter swings away
2nd   1.189 .725   x     results first stolen base second is sac bunt

here are the results.
sac bunt .725-.953   -.228 bad news
SB success 1.189-.953   .236 hooray
total swing successful steal vs. “successful” bunt   .464
flip side
SB failure vs. sac bunt
SB failure .953-.297   -.656
bunt -.228  
.228-.656 -.428

so if i did it correctly then a steal of second base is worth .464 runs over a bunt
and a failure to get second is worth -.428 runs over a bunt

so does that mean that 2 steals instead of sac bunts equals a run a game you tell me.

One more thing I just completed the math and you would have to steal successfully at least 81 percent of the time to make the risk worth it and no one steals successfully more than 81 percent of the time.  Check Ricky Henderson SB 1,406 CS 335 SB% .808