At the Blue Rocks home opener tonight (which I left early because the wind made it feel so cold), I sat near two guys who were complaining about the state of baseball today. One complaint they had was that many starting pitchers don’t enter games with the intent of completing them the way they did back in the 60s and 70s. While I tend to agree with this notion, I do think there are good reasons for it. There are arguments that pitchers are throwing many more breaking pitches nowadays which put great strain on the arm, especially as compared to fastballs. As a result, pitchers get tired out more quickly. Also, what empirical evidence exists seems to suggest that pitchers are throwing more pitches per batter faced than in the past, and offense is at a higher level than in the 60s and 70s, so they’re facing more batters per inning as well. As one example of this, in his perfect game in the 1956 World Series, Don Larsen required only 97 pitches. In tonight’s Phillies game, Brett Myers threw 88 pitches in only 3 and one-third innings.
But they saved their greatest bile for the decrease in sacrifice bunting saying that players don’t bunt the way they used to. They even argued that a cleanup hitter should be bunting with a man on first and nobody out. There’s a very simple reason that fewer players bunt nowadays. Statistical evidence shows that the sacrifice bunt is almost always a losing proposition.
Before getting to the statistical analysis of this proposition, let’s reason through it. There are only three outs per time at bat. Once a team makes three outs, all their offensive progress is erased. As a result, outs are extremely valuable things that should only be given up when absolutely necessary. A base can be gained through a number of methods: stolen base, hit, walk etc. Giving up an out is rarely worth that.
I’m am actually a fan of the sacrifice bunt. I think it makes the game more exciting then the “take-and-rake” approach many modern sluggers use. Unfortunately, the wait for your pitch approach is more effective in producing runs in the current run scoring environment.
The following table is taken from the Baseball Analysts study Empirical Analysis of Bunting:
AL 0 1 2 NL 0 1 2 ----------------------------------------------------------------- --- .498 .266 .099 --- .455 .239 .090 x-- .877 .522 .224 x-- .820 .490 .210 -x- 1.147 .693 .330 -x- 1.054 .650 .314 xx- 1.504 .922 .446 xx- 1.402 .863 .407 --x 1.373 .967 .385 --x 1.285 .907 .358 x-x 1.758 1.187 .507 x-x 1.650 1.123 .466 -xx 2.009 1.410 .592 -xx 1.864 1.320 .566 xxx 2.345 1.568 .775 xxx 2.188 1.487 .715
A brief note on reading the table. First and fifth columns are runners on base, an x representing a man on the base corresponding to the position the x is in. So, “x–” represent man on first, with second and thirds empty. Then three columns after each of those represent the average runs scored in that situation with 0, 1 and 2 outs respectively. Columns one through four are the American League with 5 through 8 the National League and this study covered the years 1997 – 1992. So, with the bases loaded and no one out, the American League averaged scoring 2.245 runs each time that occurred. As you can see from examining the table, there is no situation improved the average runs scored.
Next, they broke the bunt strategy down by batter position in the lineup. (It’s too long to reproduce here, you really should read the whole article at the link above.) With this closer examination, they found only one time that the bunt increased the average runs scored: when the National League’s 9th place hitter (usually the pitcher) was at bat with men at first and second with no one out. Even with a man on first, no one out and a National League ninth place hitter up, the sacrifice bunt is a bad idea. (Now, this still likely means that with the pitcher up in this situation, the bunt is a good idea since the difference is small enough that it would be changed if they separated out non-pitcher plate appearances in this situation.)
They then look at the results of the impact by examining run averages when managers bunt versus when they don’t and the study shows that managers are judicious enough in their strategy to minimize the negative effects of bunting, but those effects were still present.
Finally, they look at how successful the bunt is in getting the offense one run, even if it does decrease the odds of getting a “crooked number” in any one inning. There are time times this makes sense as a strategy. For example, late in the game it might make sense for a team with a one run lead to get a second run as it is more than twice as hard to score two runs as it is one. Or a team might try to get a single run to force a game into extra innings, rather than go for broke to win it in nine innings. There’s a table that shows there are a number of situations where this one-run strategy does in fact pay off.
One caveat about this study: the negative effects of the bunt in today’s higher offensive game are probably understated as with fewer runs scored over of the years of this study, the out caused by the sacrifice bunt prevented fewer runs from scoring. Although it’s an open question to my knowledge how much of this increase in offense may be due to the decline in the use of the sacrifice bunt.
So, in general, eschewing the bunt and going for a multiple run inning is a winning strategy. This really shouldn’t be a surprise to anyone, though. After all, when did God create the Heavens and the Earth but in the big inning?