Sunday, August 28, 2005

Best of the Bunt

Basil over at Nationals Inquirer helpfully points to Dave Sheinin's piece in the Post analyzing the utility of the sac bunt. It is a good article, worth reading to get a summary of the various positions. Like most pieces in traditional media, it does not go far enough. Like other bloggers, we've been following this issue for a long time, and we've placed on reserve here some additional reading materials for those interested in studying the issue further.

For a response to Bill James's position, see here. For numbers to support my position, see the links here. I don't understand why people haven't called James on this position. It would be easy to do.

For a chart showing why the "one-run" strategy is not really advanced by the sac bunt, see here.

For a case study of Frank's view, see here.

Bottom line: You should bunt when you want to preserve your chance of scoring the lead runner for the next batter. So the decision rests on whether the current batter can advance the runner by swinging away and avoid a DP. I don't mind the bunt by most pitchers -- they will often squander the scoring chance if left to swing away. But if you can pinch hit for the pitcher, you should do that rather than bunt. And most position players should never bunt. But, as with all things, the circumstances might dictate a different decision.

Update: Sheinin's game story quotes Jose Guillen explaining his inexplicable bunt in the fourth inning yesterday this way: "I was just trying to make something happen." This is the real danger of Frank Robinson's approach to the bunt described in the Sheinin piece. It is based on the notion that the bunt "makes thing happen". This is a critically important point: THE SACRIFICE BUNT DOES NOT "MAKE THINGS HAPPEN." IT SIMPLY BIDES TIME UNTIL A BETTER HITTER COMES ALONG TO "MAKE THINGS HAPPEN" -- BY SWINGING THE BAT. Bunting for a hit can "make things happen", to be sure, but not with runners on first and second. The problem is that Frank's uninformed view that bunting is a proactive play filters down to guys like Guillen who abuse it even worse than Frank.

13 Comments:

At 10:40 AM, Blogger Basil said...

I think that last sentence is the truly relevant one. James makes a lot of sense when he discourages the play but encourages a multi-layered situational look at it. For instance, it makes a lot of sense to put the pressure on a young third baseman, or one who has a reputation for wide throws.

 
At 11:18 AM, Blogger DM said...

I see the point about the "young third baseman" but the problem with it is that you are not really employing a sacrifice bunt in that situation -- you are bunt for a hit or error, really. In my earlier post I noted that James' points depends on the play going "wrong" in favor of the batting team. I don't think that's a case for the _sacrifice_ bunt, it's a case for the bunt.

 
At 12:12 PM, Blogger Basil said...

Well, it depends on the skill of the bunter. I had actually envisioned the SAC bunt in my comment, and I don't think James is conflating SAC chances from more surprising or opportunistic uses (though I could be wrong).

Sure, if you treat the SAC bunt as a totally pro forma endeavor (bunt safely and soundly, take you out at first), it's a loser play a lot of the time. (I think Ball-Wonk provided a thrilling exegesis on this earlier today.)

But, if it is employed with something beyond rote skill, if it is plopped down in something other than an obvious spot, if the third baseman is rushed or the pitcher/first baseman are forced into confusion, the opportunity for success independent of an out given away presents itself. Not a tremendous opportunity, but a tangible one. I think that's James' point---even for a sacrifice situation.

And, truth be told, it's a debatable point. The SAC bunt is a "fundamental" play because of its utility as an object (sacrament?) of sacrifice. There are attendant consequences for getting too fancy, like bunting into two strikes, popping up, or poking it straight at a fielder for a DP.

But, I think that if players were taught to treat it as a bit more of a weapon than as a craft, it could be more of something like James envisions. Is it worth it, even then? I don't know.

In principle, though, I think it should be discouraged but treated situationally nonetheless.

 
At 12:27 PM, Blogger DM said...

I am all for batters who have the skill and judgment to use the bunt to put pressure on an infield, even in obvious sacrifice situations. Brett Butler was a master at that. Ideally, that would result in the infield/third baseman playing in more on that batter, who could then exploit the real tangible opportunity for damage that such a defense presents.

My point is simply that ain't why Frank or any of the other managers are using the bunt, and that ain't what the statistical analysis against the bunt is trying to address. So James' defense is misplaced at best. Plus, the Dan Agonistes's numbers linked to in my posts seem to quantify the "tangible but not tremendous" opportunity James' identifies as not very significant.

 
At 2:12 PM, Blogger Mean Dean said...

A post I made elsewhere:

"I understand Bill's argument about how we can't just evaluate the bunt as a play by comparing the run probability of (runners advance 1 base + 1 out) to the current run probability; that it's more complex than that. It's a point well taken... but he has been making this same argument since his book about managers, which was probably at least five years ago. Is it really that difficult to find an actual answer? Look at what actually happens on bunt attempts; which results happen what percent of the time? Compare the results of similar situations in which the bunt is used vs. when it isn't; how many runs are scored on average, and how many times does the team at least score once? He makes it sound like it's an unsolvably complex problem, when it seems to me like it's just slightly more complex in an easily solvable sort of way..."

Or, to sum up, he clearly has the information available to derive the answer he claims to be looking for, so, why doesn't he just do that, instead of continuing to argue this "well, you still could be wrong" point?

 
At 2:17 PM, Blogger Basil said...

I take it that he believes it's too complicated to answer or that he has answered it (or attempted to) and his info is now considered proprietary. Who knows, but the criticism of him is indeed apt there.

DM, point taken, but I'm not sure, if I were a manager, I'd place complete fealty in a run estimation table or research conducted such as that to which you've linked. Not because I disbelieve it, but because every play is unique and I think that an occasional play, while not a good gamble on average, might be a decent gamble in isolation.

 
At 2:19 PM, Blogger Basil said...

But you have me here:
My point is simply that ain't why Frank or any of the other managers are using the bunt, and that ain't what the statistical analysis against the bunt is trying to address. So James' defense is misplaced at best.

It's true, although it's frustrating to me that I bunt is simply such a relaxed, resigned play. I wish SAC bunts were employed in a more daring manner by accomplished bunters from time to time.

 
At 2:25 PM, Blogger Studes said...

Well, James Click already did a great job of answering the question at Baseball Prospectus. It was a three-part series, but here's an incredibly simple statement he made at the end:

"Therefore, in the broadest conclusion possible, we can say that sacrificing is a good idea when pitchers are batting and, for most of the hitters in the league, when there is a man on second, no one out, and a single run is the goal."

The fuller answer is much more complex. Here's the URL:
http://www.baseballprospectus.com/article.php?articleid=2869

Also, Tom Tippett weighed in with this article a while ago:

http://www.diamond-mind.com/weblog/2003_08_17_archive.htm

(scroll down a bit to find his comments).

 
At 2:29 PM, Blogger DM said...

Basil, I think we agree on the point that inflexible rules are bad regardless of the way they are applied (bunt or no bunt). They way I would do it is use the research to develop criteria or questions you should ask yourself in making a decisions, like these I laid out in my post from May:

(1) Can I afford to limit this inning to one run?
(2) Is the batter likely to strike out or pop-up?
(3) Is the batter likely to ground into a DP?
(4) Is the third baseman a bad fielder, likely to throw the ball away or make a dumb play?
(5) Is the batter fast? Might he beat out the throw?
(6) Is the batter able to get the bunt down?

Note that these questions avoid the fallacy that trading the out for the base improves your chance of scoring. The focus is no longer on the base advance (where it shouldn't be) but on the characteristics of the batter and fielders, where it is best placed.

 
At 2:31 PM, Blogger DM said...

Hey, Studes, thanks for joining the discussion and for the links.

 
At 2:35 PM, Blogger Basil said...

DM, that's a great checklist. I think your approach would make sense. I'm sort of babbling in this thread, probably in an area I shouldn't be, but I don't want to give the impression I would bunt willy-nilly. Far from it, and far, far less than Frank does. Okay, that last part is stating the obvious! ;-)

Studes, obviously that research represents very good work in the field, and it's worthy of great respect. I would only imagine (hope?) that if James has something solid to contradict it, he's in a situation where he can't share it. And, yes, I know it's the same line he's been using since the managers book, which I read in '97, if not earlier.

 
At 4:24 PM, Blogger Studes said...

Should have read the article before giving the links. I didn't know Click was quoted so extensively. That's a good thing.

Regarding James, he analyzes what he wants and doesn't feel a particular responsibility to come up with all the answers. I actually doubt he's delved into the detail of this one.

 
At 11:32 AM, Blogger DM said...

Basil, as I thought about your points some more this morning, it came to me that what needs to be done is pitchers (in the NL) are the prime candidates for the aggessive bunting instruction, beyond the pro forma bunt you describe. I've always thought more work could be done by pitchers to improve their hitting, and maybe the bunt is the place to start that more work.

 

Post a Comment

<< Home