Brewers' bunting

[jeffy]

My understanding is that we should be using something similar to this instead of basic base out states. After all, you play to win the game.

According to that, with the home team down one run in the bottom of the ninth, they are better off with a runner on 1st and no outs that with a runner on second and 1 out. 33% chance of winning in the 1st scenario 28% in second (I used 4.5 run scenario, assuming that is about average). This too is overly simplified as it assumes a successful bunt, while on average the success rate i something like 70%, IIRC.

So the bunt scenario is really a ~70% chance of a runner on second with one out and a 30% chance of a worse situation such as a runner on 1st with 2 outs or bases empty and 2 outs. The 70% is not quite right either as there is a chance the bunt results in a hit or an error.

Having read "The Book", as others have mentioned, what the defense is doing is a big factor. Bunting when they are not playing for it can be a good strategy, bunting when the 1st and 3rd basemen are about even with the pitcher's mound, not so much.

[TooLiveBrew]

I know the payoff is big, but when you walk a guy to 'set up the double play', you're betting on a scenario that, even for a slow runner, occurs at most roughly 30 times out of 600 PAs... playing that 5% chance!

That is not fair because you can only ground into a DP when you come to the plate when a DP is in order with less than 2 outs, and you have far less than 600 PA is those scenarios so the DP is much, much higher if you only consider PAs with a DP in order. There are many balls hit during the season by a player that would have been DPs if there was a runner on first.

 

That's a good point. I'm glad it's Friday.

[logan82]

According to that, with the home team down one run in the bottom of the ninth, they are better off with a runner on 1st and no outs that with a runner on second and 1 out. 33% chance of winning in the 1st scenario 28% in second (I used 4.5 run scenario, assuming that is about average). This too is overly simplified as it assumes a successful bunt, while on average the success rate i something like 70%, IIRC.

So the bunt scenario is really a ~70% chance of a runner on second with one out and a 30% chance of a worse situation such as a runner on 1st with 2 outs or bases empty and 2 outs. The 70% is not quite right either as there is a chance the bunt results in a hit or an error.

Having read "The Book", as others have mentioned, what the defense is doing is a big factor. Bunting when they are not playing for it can be a good strategy, bunting when the 1st and 3rd basemen are about even with the pitcher's mound, not so much.

Yes you still have to factor in success rates of a sac bunt but it gives you a much better indicator of when you will actually improve your chances of winning than just base out states. When something lessens your chances of winning even if you are successful why keep doing it?

 

The bolded portion is probably the most important part. The only time bunting should be done is when the defense will not expect it or is not playing for it. Bunting while the defense is specifically playing for a bunt is a really bad idea.

[TheRealMattKain]

topper09er]

TheRealMattKain wrote:

Is summary:

When you bunt and move a runner over at the cost of an out, historical data shows that you are less likely to score 1 or more runs in the inning. Your chart very clearly proves this.

No, you are less likely to score 1 run, you are more likely to score more than one run.

 

What exactly are you saying "No" to? Please tell me what I wrote that you think is wrong. I repeat, (based on the data that you presented) with no outs and a runner on first you are more likely to score a run (any number) than with 1 out and a runner on second. Are you disputing this? Do the math - add up the percents in the cells. Who cares if you're less likely to score just one run? Since when do you get penalized for scoring more runs? In all of the other cases that you're ignoring (scoring 2 runs, 3 runs, 4 runs, etc.), not only did you score one run, you scored more! If you're down by a run then scoring 1 run is nice and scoring 2 runs is even better, and scoring 3 runs is even better, etc. If you hit a homerun with the bases loaded the odds of scoring just 1 run are 0%. So you should never hit a homerun with the bases loaded? Of course not, because if you hit a homerun with the bases loaded the odds of scoring 4 runs is 100%.

 

Now I agree, that it's far more complicated than the chart that you posted, but please help me understand. Based on your chart, are you disputing that with no outs and a runner on first you are more likely to score a run (any number) than with 1 out and a runner on second?

 

 

[Baldkin]

Yes, in looking at the chart: 1b and 0 outs: 56% of the time = 0 runs.

2b and 1 out: 59% of the time = 0 runs.