<$BlogRSDUrl$>

Talking Baseball

Your weekday baseball fix. Some days.



Posted by Dave on Wednesday, April 28, 2004

Why Are We Not Sacrificing the Sacrifice?

I recently got an e-mail from some devoted readers. These weren't just any devoted readers, however. These were the readers of paramount importance, at least socially. These were my girlfriend's parents. Seriously though, they're both wonderful people and are inquisitive minds. Their curiosity and interest in baseball led them to ask me about the practice and frequency of bunting in a recent game between the Red Sox and Yankees:
To Bunt Or Not to Bunt? That is the question...We are watching the redsox yankees game and getting frustrated with the bunts (and sacrifices in general). We believe that they are OUTS!!!! and therfore not good...What is your opinion??
- Cathy and Thom
Debo’s parents seem to have the correct intuition here. By giving the defenders outs, a team is worsening their chances of scoring runs. After Cathy and Thom sent the inquiry, I responded with a rather informal and uninformative e-mail. I basically said there were many studies that have been done that show that sacrificing is generally an ineffective strategy as compared with hitting away. Did I cite these studies? I did say the e-mail was uninformative. Though I was too lazy and too tired to include the studies in my e-mail back to Cathy and Thom, you can read about the dubiousness of sacrificing. Though I couldn't tap the source - the article written by Tom Tippett over at Diamond Mind - you can read about the results of the article. A biased sabermetric opinion would tell you that sacrifice bunts are always worthless, but that's not the case. These bunts are just usually a bad idea. When is it a good idea? A runner is on second with none out and Aaron Rowand - check that, Aaron Heilman - is at the dish facing Eric Gagne. With such a weak hitter and such an over-powering pitcher at the plate, it would make sense to bunt in only these sorts of situations. You can find a temperate view of sacrificing expounded by Tippett (with commentary from Neyer) here.

Even though I had given some explanation, Thom was still dissatisfied and uncomfortable:
If it is obvious [that bunting is generally a bad idea], why do managers keep doing it? During this last Red Sox-Yankees series we squandered a few men on-base to this strategy. I am wondering if it was viable when players knew how to bunt (perhaps 40 years ago) and it remains embedded in the minds of the current managers.
- Thom
Thom poses a fantastic question. If the knowledge is available proving that bunting is generally an ineffective strategy, why on earth would managers continue to command their underlings to behave in a non-optimal manner. Remember, the study says that - in almost all cases - bunting worsens a team's ability to score even one run - let alone many. The possibilities of why managers would forego the optimal strategy (that is, to not sacrifice) are many, let's take a survey of them:

1. Managers could merely be unaware of the empirical proof of the futility of sacrifices. Diamond Mind is not the preeminent baseball website (that honor probably belongs to MLB.com or ESPN.com's baseball section), but baseball men should be reading it. It is the manager's job to manage his team in the most effective manner, and this should imply that they will always be up to snuff on the latest theses about empirical evidence. I suspect, however, that distrusting, traditionalist personalities would avoid the findings (such as Dusty Baker, perhaps (Chalk it up! (on the counter (Sorry, I'm too much of a math-nerd to avoid the temptation to do this (having millions of parentheses))))). This bleeds into another possibility...

2. Managers could actively ignore and/or refuse to accept the evidence. I can almost sympathize with the managers in this case. Surely, not many (any?) managers are econometricians, statisticians, or economists (though, in an ideal world, they should all have such a background). This renders the illuminating evidence as incomprehensible. Worse, the evidence is contrary to the baseball "truths" that have been propounded to them as they matured into a manager. What are they going to trust? A bunch of facts they can't wrap their minds around, or the "truths" that've been preached to them since childhood? I can't blame them for trusting their compatriots. Fear of the unknown, after all, is natural. For a manager, however, it is not forgivable. As a manager, it should be your duty to make yourself of the methods used by economists in order to understand their arguments. It is your job to strategically optimize how your team plays. If this means you have to learn how to do some regression analysis on the side, so be it.

3. There may be a fear of deviating from standard operating procedure on the manager's part. This is still a real fear, sadly. Suppose a manager declines to bunt in a conventional situation (runner on first, no outs, tie-game). Since the members of the media are still, by and large, ignorant of the objective revolution and sabermetrics in general, the media will trash the manager if his non-bunting strategy fails. The media, unfortunately, heavily influences public opinion. With the media trashing the manager, the public will sour on the manager. Since upper management and ownership responds most strongly to public opinion, the possible malcontent with the manager increases his risk of being fired – even if he was doing his job better than Dusty et al (Count it!). What if the manager had used a sub-optimal strategy? Even if it had failed (which it is more likely to do), the custom and convention of baseball would have absolved him of utilizing an incorrect strategy – the team failed to get down the bunt, the writers would say. Worse, even if the player sacrificed successfully, the writers would laud the move. Unfortunately, as Tippett found, even if the sacrifice succeeds a team is still lowering its expectation for even one run. I can’t wait for the day when a progressive writer questions a conventional bunt in a mainstream newspaper.

4. The penultimate reason a manager may continue to bunt in poor situations is because he maintains a degree of control over the proceedings. A manager’s job is very hands-off – especially with respect to in-game management. This is especially true following the sabermetric revolution. Many proactive managerial decisions – to steal, to frequently change relievers, and now, to sacrifice – have been shown to be sub-optimal strategies in most situations. A manager may want to feel like he has a great degree of control over the game’s proceedings, so he may call for a bunt or steal more often than is necessary. A manager would be motivated to signal for sacrifices because his proactivity would signal that his job requires action – that managers need to act to do a good job. Sometimes, however, inaction is better than action.

5. Lastly, sacrificing brings tangible rewards. By sacrificing, a player will nearly always move up an extra base – generally moving into a crucial position. It would seem that this benefits the team by accomplishing something nearly all the time. After all, the majority of the time (60% of the time – or more) if a hitter swings away in the same situation, it’s very likely he’ll record an out that is wholly unproductive. Because managers are generally risk averse (even though they should be risk neutral, of course). For a discussion of risk aversion, scroll to the bottom of this article. Because they’re risk averse, they discount potentially risky situations more. Swinging away induces more risk in the situation - the at-bat could be a total bust, but it could also be much more successful than a sacrifice. If we just look at the expectation (this implies the manager is risk-neutral) of just one runs or multiple runs, however, we should generally choose to swing away. Since we’re reducing our expectation for run-scoring, sacrificing is a sub-optimal strategy. But, since the managers are risk-averse, they’ll value secure benefits - that station advancement for the base-runner.

So, why does superfluous bunting still exist? Well, clearly it’s a product of some combination of these five possibilities. Are there more reasons for the presence of an overabundance of sacrificing? Certainly. Let me know what you think by commenting below.

He’s More Effective Because of His Increased K’s…I Mean…Z’s

The artist formerly known as Ismael Valdes used to be a fairly ordinary pitcher. A lifetime ERA of nearly 4.00 would label anyone mediocre. But Ismael knew something was amiss - he never felt quite right. Finally, prior to this season, he identified the problem. He was mislabeled his entire life - not as a mediocre pitcher, but as a person! It turns out Ismael’s real last name is Valdez, not Valdes. He changed his name prior to the season, and his increased comfort is already reaping tremendous benefits. Through 22.3 IP, Valdez has recorded a 2.42 ERA.

I was always just on the cusp of athletic stardom, and, this just in, my birth certificate reports that my name is David Mets. Figures.


### So what do you think? We want to know. | | E-mail us ###