CHAPTER 9 — TO SACRIFICE OR NOT

Few strategies elicit as much emotion and controversy as the sacrifice bunt. In 1990, Jay Bell of the Pirates had 39 sacrifice hits. He presumably had many more sacrifice attempts than that. Some people would argue that he pre-emptively killed a plethora of Pirate rallies, while others might argue that his skill as an accomplished bunter created more runs and more wins for the Pirates than if he had swung away instead. As is often the case in baseball, fans and pundits can argue such things until they are blue in the face, but without a detailed and thorough analysis of the numbers, there is simply no way of knowing wherein the truth lies.

Such rhetoric as, “It is a good strategy to put a run on the board as quickly as possible,” or, “Late in a close game, getting a single run can be critical,” or, in the opposite vain, “Outs are too precious to waste,” or, “A sac bunt is a rally killer, period,” are just that—rhetoric. They don't enable us to answer the following pertinent questions:

• Is it ever correct to sac bunt? In other words, does a sac bunt ever increase a team's chances of winning the game, as opposed to swinging away?
• If yes (it is sometimes correct), when?
• From what positions in the lineup, by what kinds of batters, with what kinds of hitters following the batter, versus what kind of pitcher, in what park, at what score, and in what inning?

If we can answer all, or even some, of these questions, then perhaps we can put the controversy and the arguments to rest. Maybe.

Looking at historical data, we see that the sacrifice bunt is a strategy that has dramatically fallen out of favor over the last 25 years or so. In the 1940's, the sacrifice bunt rate in the NL and AL (for non-pitchers) was around 5.5 per 500 PA. By the 1980's, that had dropped to around five, and in 2004, it was only a little over three. Why is the sacrifice bunt not as popular as it once was? As with the stolen base (which has also declined dramatically), many GM's and managers have come to the realization that small-ball strategies, like the stolen base and sacrifice bunt, are not as effective in the high run environments prevalent in the modern era of baseball.

As well, conventional sabermetric wisdom says that the sacrifice bunt is generally an ineffective and archaic strategy. For example, the three highest-profile sabermetric teams in 2004 in the AL were Oakland, Boston, and Toronto. While the AL average for sacrifice hits per team was almost 39, Boston had only 12, Toronto 10, and Oakland 25, the three lowest totals in the league. The sacrifice bunt appears to be a strategy eschewed by sabermetric teams.

Is sabermetric conventional wisdom correct? (Note that you will not often see the word sabermetric juxtaposed with the term conventional wisdom, at least not in this book.) You may be surprised at the answer!

Where does the sabermetric notion that a sac bunt is rarely (if ever) productive come from? Well, way back in 1984, Pete Palmer and John Thorn, in their groundbreaking book, The Hidden Game of Baseball, said, “The sacrifice bunt ... is a bad play.” They go on to say that, “With the introduction of the lively ball, the sacrifice bunt should have vanished...” They are talking about non-pitcher bunts of course. In fact, unless otherwise noted, throughout the rest of this chapter, when we speak of a sacrifice bunt or just a bunt, we are speaking of a sacrifice bunt or attempted sacrifice bunt by a non-pitcher.

Where they derived their conclusions, as outlined in The Hidden Game, was from the run expectation and win expectation tables that we have constantly been referring to in this book. Here is a copy of Palmer's original run expectation chart modeled after the 1961–1977 baseball seasons.