Settle this Baseball Argument
 

It is the bottom of the 9th in a tie game. You have men on 1st and 2nd with no outs and your 2,3,4 hitters coming up to bat. The man on 2nd has slightly above average to above average speed. Your 2,3,4 hitters are your best 3 hitters, can all hit for power and all have solid BA/OBP numbers, and are all significantly better hitters than the #5 hitter. What do you call for with the #2 hitter coming up to bat?

Please answer the poll question and explain your answer in thread.

Re: Settle this Baseball Argument
 

The answer can be determined factually with run probability. I'm too lazy to do it, though.

Re: Settle this Baseball Argument
 

Ok, take a look at the Run Frequency matrix, for a start. With runners on 1st and 2nd with no one out (the current situation) you will score 1 or more runs (the only thing we care about) 64.1% of the time. We have 4 possible outcomes. The hitter swings away and gets a hit (we'll assume for simplicity that ALL hits score him), he could swing away and get out, he could bunt and succeed, or he could bunt and fail. Lets take a look at these outcomes and their impact on how many runs we score.

Swing and Hit: Win the game.
Swing and get out: 42.6% to win if runners stay, 69.5% to win if runners advance, 26.3% to win if he GIDP.
Bunt and succeed: 69.5% to win
Bunt and fail: 42.6% to win.

So, now you need your assumptions. What is the BA for the 2 hitter? What is his successful bunt percentage? Take guesses at these, and you can quickly figure out what your EV is in each choice. If he is a .300 hitter, and he GIDP 10% of the time, and he hits a deep enough fly ball another 10% of the time, you have .30(1)+.10(.263)+.10(.695)+.5(.426)=61% to win if you swing away. So, if the guy gets the bunt down 100% of the time, bunting is obviously best. But he probably doesn't. So, whats the breakeven point? In this case, it would be about 88%. If he successfully bunts 88% of the time, bunting is the better choice. With all of these assumptions. If I did the math right (another huge assumption). Do most good 2 hitters beat this 88% mark? Probably, but I bet its close. Like in poker, this is probably a decision that just doesn't matter. Go with the one that doesn't get you fired when it doesn't work (i.e. the bunt).

The only thing that the abilities of your 3 and 4 hitters matter for is that you can upgrade your run expectancies a little bit in ALL scenarios. This means you can probably ignore all of that, although it isn't entirely immaterial.


Edited in some stuff

Re: Settle this Baseball Argument
 

We can use imaginary Derek Jeter.

.318 when he rakes we walk off...
.071 when he swings we load the bases...
.02 he grounds into a DP

We can assume best case 100% bunt success. And say that half the time there is an out runners advance.

.318+.071(.894)+.305(.426)+.305(.695)+.02(.263) = 72.8%

So if Jeter were to be 100% bunter, it would still like be +EV for him to swing unless the person fielding the bunt was McKlutz and would have an error >3%

If you wanna jazz this up, find a better bunt rate and a how often a runner advances to 3rd.

Re: Settle this Baseball Argument
 

Bunt is right here, I think. RunFreq goes up while RunExp goes down.

EDIT: This is way too simplistic. Refer to the above two posts for better analysis. I already thought way too much about this in my post.

Re: Settle this Baseball Argument
 

[ QUOTE ]
Bunt is right here, I think. RunFreq goes up while RunExp goes down.

[/ QUOTE ]

Thats what I got with my assumptions, I'm not sure what Thremp did differently (he used different assumptions a little but I dont think that is it?) maybe he was using RunExp or something like that, i.e. not accounting for the fact that 1 run = 5 runs in this spot.

Re: Settle this Baseball Argument
 

[ QUOTE ]
[ QUOTE ]
Bunt is right here, I think. RunFreq goes up while RunExp goes down.

[/ QUOTE ]

Thats what I got with my assumptions, I'm not sure what Thremp did differently (he used different assumptions a little but I dont think that is it?) maybe he was using RunExp or something like that, i.e. not accounting for the fact that 1 run = 5 runs in this spot.

[/ QUOTE ]

No, I pretty clearly used WinExp. Keep in mind Jeter is a pretty good 2 hole hitter. Your breakeven point in reality might be closer to a .290/.370 type guy. My biggest worry about what I did is how often runner's advance on outs.

Note: I always bunt v Tigers. This include in the first inning with Pronk up to bat.

Re: Settle this Baseball Argument
 

They bunt because Willie Randolph is the team's manager.

Re: Settle this Baseball Argument
 

Bunt 50% of the time.

Re: Settle this Baseball Argument
 

Double Steal, the guy will be safe on 3rd, and the 3rd basement will throw the ball into the outfield (trying to get the guy at 2nd out) and thus allow the runner to score.

<font color="white"> in reality, i would say swing away assuming he is a good hitter like in the op </font>

Re: Settle this Baseball Argument
 

Hardy, Braun, and Fielder are not "significantly" better than Bill Hall. In fact, Hardy is not at all a better hitter than Hall.

I believe bunting was the correct move in each case last night. The next hitters just came up with the absolute worst AB's they could have.

Re: Settle this Baseball Argument
 

Swing away. You didn't say whether or not #2 is good at bunting, so I assumed average.

When bunting, you can pop it up to catcher, or even get the guy out at third, or possibly if bunted too hard to pitcher, get a 3rd-to-1st DP. There is a lot that can go wrong. A successful bunt will allow #3 the sac fly option, though.

Swinging away may cause double play, but most likely would leave guy at 3rd who can score on any hit from #3 hitter. Striking out would suck, but whatever. Also, many singles will score runner from 2nd base (although 3rd base coach likely holds them up on close play as not to make first out at home).

And you give #2 hitter chance to be hero (See: Mr. November/Captain Clutch), which is more important to owners than winning. [img]/images/graemlins/grin.gif[/img]

Re: Settle this Baseball Argument
 

[ QUOTE ]
Hardy, Braun, and Fielder are not "significantly" better than Bill Hall. In fact, Hardy is not at all a better hitter than Hall.

I believe bunting was the correct move in each case last night. The next hitters just came up with the absolute worst AB's they could have.

[/ QUOTE ]

Braun and Fielder absolutely are wayyyyy better hitters than Hall, the only person you could even argue against would be Hardy.

My non-statistical way of thinking about this question was this. With your best 3 hitters coming up and only needing a single to win, except for the slight possibility of a double play, if you have them swing away, the opposing team will need to get through all 3 of your best hitters to prevent you from winning the game. Bunting in this situation effectively takes the bat out of 2 of your 3 best hitters. The #2 hitter now records a basically automatic out, and there is always the possibility of an unsuccessful bunt that results in a pop out or a force out at 3rd, or two foul bunts in a row that result in having to swing away with an 0-2 count. Now even if the bunt does succeed you have men on 2nd and 3rd with 1 out, and the opposing team can choose to walk your #3 hitter and pitch to the #4 hitter with the bases loaded, so now you've lost the ABs of your 2 and 3 hitter. Or they can pitch to the #3 hitter and if they get him out, walking the #4 hitter becomes obvious. Basically the point is, if the 2 hitter bunts, there is no way the 3 and 4 hitters are both getting an AB unless the bunt was unsuccessful. So now instead of having to deal with all 3 of the team's best hitters, the opposing team effectively only has to get out 1 of your best 3 hitters, and gets their choice of 1 of 2 to face.

Re: Settle this Baseball Argument
 

The success of a sacrifice attempt varies based on the D.

If you have The Book, check out page 283 (first edition). Long story short, its much harder to bunt in the 9th inning. To quote "The net result is that an average batter who bunts (as slightly worse than a league average hitter) achieves a greater WE for hitting away than from attempting a sacrifice, even in the bottom of the ninth in a tied game. Without going through the numbers, the results are similar with a runner on second or on first and second."

If your 2 hitter is average, Jack is right. It sounds like this 2 hitter is above average:

"In the ninth inning of a close game, below average hitters should bunt most of the time, given typical speed and bunting ability. Average hitters should bunt about half the time. Above average hitters should rarely bunt. The speed of he lead runner appears to be a significant in the success of failure of the sacrifice attempt in the ninth inning, and as always, the speed and bunting proficiency of the batter should be strongly considered." (p.284)

DP's matter a lot too.
"With a runner on first or first and second, and no outs, the batter's GDP rate (adjusted for the pitcher) should be considered in deciding whether to bunt." p 280
The

In a tied game in the bottom of the 9th, does not WE = RE?

Re: Settle this Baseball Argument
 

Mostly hit away. Fast runner on second, #4 hitter who doesn't hit into a lot of double plays, bunt more often.

Re: Settle this Baseball Argument
 

Augie Garrido, who has won more games than any other college coach ever, said that with 1st and 2nd and no outs, he would bunt Babe Ruth.

Re: Settle this Baseball Argument
 

I assumed:

.300 hitters

.7 chance of successful bunt

.6 chance of hit or SF

I got 64% and 65% chance of scoring 1 run for bunt/no bunt.

Re: Settle this Baseball Argument
 

Then he's a goddamn idiot.

Re: Settle this Baseball Argument
 

One of the few situations where I think a bunt is absolutely correct. But I am assuming that the 2 hitter is a competent bunter.

And I really have no idea what the stats are for successful sacrifice bunts. Generally, what % of attempted sacrifice bunt attempts fail? And what % result in infield hits?

Re: Settle this Baseball Argument
 

Umm... It seems that above average hitters should never bunt here.

Re: Settle this Baseball Argument
 

Yeah.. I admit to being a baseball stats noob, but JoA's suggestion of bunting 50% of the time doesn't make much sense to me.

Re: Settle this Baseball Argument
 

Butcher boy, FTW.

Re: Settle this Baseball Argument
 

[ QUOTE ]
Yeah.. I admit to being a baseball stats noob, but JoA's suggestion of bunting 50% of the time doesn't make much sense to me.

[/ QUOTE ]

I think it is to keep the defense honest and make the 3b play in which increases the chance of a hit next time. Bunting some of the time will make them come in next time, where you dont bunt and have a higher chance of a hit than normal. Bunting will sacrifice some EV, but since it usually gets the runner over to 3rd, what you are giving up is small enough that it is more than made up for by the increased chance of a hit next time. I think.

Re: Settle this Baseball Argument
 

is swinging away but try to hit it to right field an option?

Re: Settle this Baseball Argument
 

[ QUOTE ]
Yeah.. I admit to being a baseball stats noob, but JoA's suggestion of bunting 50% of the time doesn't make much sense to me.

[/ QUOTE ]

Meta-game!

If you always bunt here, you hurt your chances of successfully bunting. If you never bunt here, you hurt your chances of successfully swinging away.

Maybe the percentage should be something like bunting 25% of the time or whatever. How often you bunt is really a sliding scale based on the skill of the 3B and how far in he's playing.

Re: Settle this Baseball Argument
 

[ QUOTE ]
[ QUOTE ]
Yeah.. I admit to being a baseball stats noob, but JoA's suggestion of bunting 50% of the time doesn't make much sense to me.

[/ QUOTE ]

Meta-game!

If you always bunt here, you hurt your chances of successfully bunting. If you never bunt here, you hurt your chances of successfully swinging away.

Maybe the percentage should be something like bunting 25% of the time or whatever. How often you bunt is really a sliding scale based on the skill of the 3B and how far in he's playing.

[/ QUOTE ]

This makes no sense. Are you assuming hitter is average for this?

Re: Settle this Baseball Argument
 

If you bunt 100% of the time, the defense can play at a depth which makes bunting -EV. If you hit away 100% of the time, the defense can play far enough back that hitting away is -EV. So, in order to have the defense play incorrectly, sometimes you must bunt, and other times you must hit away, and there is a ratio of bunting/hitting at which the defense cannot play optimally. If the defense doesn't try and play optimally, then you can lean towards one or the other.

Re: Settle this Baseball Argument
 

Not enough info to give an answer.

Re: Settle this Baseball Argument
 

A lot more butns fail then result in infield hits/errors

Re: Settle this Baseball Argument
 

[ QUOTE ]
If you hit away 100% of the time, the defense can play far enough back that hitting away is -EV.

[/ QUOTE ]

I just don't buy this, unless maybe the guy is a really sick bunter. Even if they know for sure he's not bunting, they're just playing double play depth. It's like a regular at bat. See Thremp's first post.

Re: Settle this Baseball Argument
 

In The Book, they've broken down bunt attempts with a man on first and 0 outs by early and late game situation, and the resulting win expectancy that a team would have should you bunt/hit away, basically the percent that a team will score in the inning following the bunt attempt. If the defense were playing as they were early in the game, you will win after attempting a bunt 71.6% percent of the time. If they were playing as they were in the 7th or 8th, you'll win 70.1% percent of the time. If it's the 9th, you'll win 69.6% of the time. If you swing away in the 9th, you'll win 71.5% of the time. So, clearly swinging away is better in most situations in the 9th, nothing else known about the skill of the hitter, the pitcher, the defense, etc. But, bunts occur in the 9th over 50% of the time in this situation, while early, they occur only 7% of the time. It makes sense that the defense would be playing in during the 9th and back before the 7th. Looking at the numbers, it's roughly equal to bunt early and swing late. However, if all you do is swing, then they'll stop playing you to bunt and it will be more like an early game situation where it is correct to bunt and incorrect to swing away, assuming you're playing for scoring a single run and not maximizing the total amount of runs scored. Therefore, sometimes you must bunt and sometimes you must swing away in order to keep the defense from playing optimally. Note that this is for only a man on first and not two men on, but I'd assume similar considerations apply.

Re: Settle this Baseball Argument
 

[ QUOTE ]
If it's the 9th, you'll win 69.6% of the time. If you swing away in the 9th, you'll win 71.5% of the time. So, clearly swinging away is better in most situations in the 9th, nothing else known about the skill of the hitter, the pitcher, the defense, etc.

[/ QUOTE ]

The flaw in the reasoning (as you've presented it) is that it was assumed that the people who bunted and failed 30.4% of the time would have failed at the same rate if they were hitting away as those who swung away and failed 28.5%-- they may have been selected to bunt because their skill set and the match up indicated a higher failure rate at swinging away. (This isn't a new thought for this thread, btw. It just indicates that that stat and argument doesn't clearly establish anything.)

Re: Settle this Baseball Argument
 

"So, clearly swinging away is better in most situations in the 9th, nothing else known about the skill of the hitter, the pitcher, the defense, etc."


I don't think it shows that it's clearly better.
It might just mean that the times they did swing-away they were correctly evaluating their greater advantage for doing so thus they won more.
This still doesn't mean that swinging-away is the default 'better' play every time this situation comes up with everything else being equal (which it never is).
It could just mean that the times when a manager DOES choose to swing-away he is more likely to have spotted something that made it more correct to do so.

IMO, you can't just look at all the times they bunted compared with all the times they swung-away and then say that in general, "swinging is better."

Re: Settle this Baseball Argument
 

I think to some extent the answer is just going to be "both options are pretty damn viable".

If I'm the M's, I am bunting, because TURBO! is a [censored] dp factory.

I also think so some extent you want to consider the nature of your 3/4 hitters. You'd rather have a guy who is going to get the ball in play with his outs a little more than a good TTO guy. This (2nd and 3rd 1 out after a bunt works) where a strikeout is really lame.

Re: Settle this Baseball Argument
 

Sometimes it makes sense to bunt good hitters.

Re: Settle this Baseball Argument
 

[ QUOTE ]

It could just mean that the times when a manager DOES choose to swing-away he is more likely to have spotted something that made it more correct to do so.


[/ QUOTE ]

Actually Bob, if sabr data support two opposite conclusions, the one that suggests that managers are stupid is the correct one.

Re: Settle this Baseball Argument
 

I'll start by saying I hate the sacrifice bunt in general.

That being said, there are a lot of factors here not mentioned.

-How is the pitcher pitching? Did he walk either or both baserunners? Is he a ground ball or fly ball pitcher?
-How good a bat handler is the #2 guy? Can he take a strike, sell hitting away and bunt for a base hit, or will the defense charge regardless?

Control, sinkerball pitcher vs good bat handler, I still favor hit and run over sacrifice.

Re: Settle this Baseball Argument
 

[ QUOTE ]
Augie Garrido, who has won more games than any other college coach ever, said that with 1st and 2nd and no outs, he would bunt Babe Ruth.

[/ QUOTE ]

As opposed to Earl Weaver, who probably would never even consider sacrificing in this (or any other) situation.

Re: Settle this Baseball Argument
 

Braun is not significantly better than Hall. Hotter and luckier doesn't mean better.

Re: Settle this Baseball Argument
 

well if you would bunt ruth there you must be very drunk

but the problem with all of these formulas are they are not an exact science and never will be and the 2-3 % diffeerence they show one way or the other can be margin of error
I dont care what a bunch of guys with similar jeter have done over the last 100 years in that spot-
He might be slower or faster than them, a better or worse bunter etc.
The infielders might react well in this spot or you could have ty wigginton fielding the ball
The pitcher might have trouble throwing strikes after intenionall walk someone or he might have maddux like control.
The guy who will bat with the bases loaded might not handle the bat well or be capable of adjusting to the situation.
My gut would say to bunt but its going to be close to break even- but there is no way this book can be accurate with so man factors that can never be accounted for.