When To Bunt

Extracting General Lessons from Baseball Reference

When To Bunt

Getting the strategy just right is crucial to the success of your replay. This is especially true for plays that are not as common in baseball now as they were 75 years ago — like the sacrifice bunt.

We can figure out pretty easily how many sacrifice hits took place in 1949. Thanks to Baseball Reference, we can see that there were 1,408 successful sacrifice hits in 1949 — or about 88 per team:

That works out to 0.57 sacrifice hits per game, or about one successful bunt every 2 games. If you’re really curious, you can look at the all time season record, where you’ll see that 1908 featured a whopping 1.30 sacrifice hits per game.

Now, we know that sacrifice flies were mixed in with sacrifice bunts for years. Per this helpful Baseball Reference Bullpen page, sacrifice flies were generally mixed with bunts from 1908 to 1931, in 1939 for a single year, and then were ignored until they were counted as a separate statistic starting in 1954.

In other words, we can say with a high degree of confidence that there was indeed about 1 bunt every 2 games for each team in 1949.

Now — this is important primarily because these are outs. With 1,408 recorded bunts in 1949, we’re talking about 1,408 outs that need to be accounted for. It’s not a huge number of outs (again, 1908 is a vastly different story), but it’s certainly important. If we decided to never bunt, as many of the modern strategists advise, those outs would need to show up somewhere else, and we’d run the risk of throwing our overall numbers off.

The question, though, is how we implement this in our games.

Strategies

Now, there are a few ways we can make sure we bunt when we’re supposed to in our replay.

We could just bunt when the situation seems to call for it. That’s a bit tricky, though, since it’s hard to tell when the right time actually comes up. I find myself scratching my head from time to time when the computer manager bunts with good power hitters in late innings of close games that the visiting team is leading (remember, I manage the home team in every game in this project).

We could try to force ourselves to bunt exactly once every other game. The problem, of course, comes when we try to bunt in blowouts, or when our bunts turn into base hits and are therefore not counted as “official” sacrifice bunts.

We could try hunting through individual game accounts. Kudos to you if you have the hours necessary and the system in place to deduct what strategies were on an in-game basis!

Or we could play around a bit with Baseball Reference’s Batting Splits page and see if we can’t come to a conclusion.

Bunt Splits

Now — you should know that some of the splits that Baseball Reference shows you aren’t all that useful. Take, for example, batting splits by month:

It’s a little bit interesting to know that bats got hot in July 1949. However, those sacrifice hit totals tell us very little.

What’s more interesting is to look at specific game situations. And we can do that quite easily and quickly.

Rather than using raw numbers, which is what Baseball Reference offers us, I’ve downloaded the data in a CSV file and have used LibreOffice Calc to convert them into rate statistics — specifically, sacrifice hits per plate appearance.

Situations

Let’s take a quick look at sacrifice hits as a percentage of plate appearances in certain innings.

Note that I am looking at a percentage rate simply because it’s the easiest option to understand. Seeing “only” 70 sacrifice hits in extra innings doesn’t seem like much — but it feels different when you realize that over 3.5% of all extra inning plate appearances were sacrifice hits.

There are a few obvious things we can take from this chart. You’re probably not going to bunt in the first 3 innings — though it did happen from time to time. You’re more likely to bunt in the late innings of a close game. Interestingly enough, the 7th seemed to be the inning for bunting — though I think we should be careful before we read too much into that.

By the way, if you really want to see a detailed breakdown of this breakdown, you can click on one of the inning links on the Baseball Reference Splits Page. That will show you the splits for that inning disaggregated by team. Here’s what the 7th inning looks like:

Now, we do need to be careful here. It’s interesting to see that the Dodgers had more sacrifice hits in the 7th inning than any other team — but they also had more plate appearances in that inning than any other team. The rule of thumb here is that the more we disaggregate the stats as we dig down, the more we run afoul of sample size issues — and the higher the chance is that our findings don’t really mean anything.

Who Bunts?

We can also use the same offensive splits page to see who did most of the bunting in real life. Here is the breakdown by position in the batting order:

It’s pretty safe to assume that pitchers usually batted 9th, by the way.

Now, it’s no surprise that the 9th place hitter (usually the pitcher) bunted so frequently. You might be surprised, though, by how frequently the number 2 hitter bunted. I assume this is in keeping with the long-deceased offensive strategy that called for the second hitter to move the leadoff hitter up a base.

It’s also interesting to see all the bunts from the hitters in the third spot. Remember that this particular metric has quite a bit of significance thanks to the large number of plate appearances. Infrequent occurences like the relief pitcher winding up in the number 3 spot in extra innings aren’t going to greatly skew the numbers here.

Who Was Bunted Against?

Finally, we can take a quick look at which pitchers faced most of these bunts.

Now, don’t let the raw stats fool you. Bunting happened far more frequently against relief pitchers. I’m assuming this is because the relief pitchers generally were pitching in situations that tended to call for bunts.

Note, however, that there were still quite a few bunts by starters going through the order the first time.

Through these simple tools and calculations, you can get a much better understanding of when it was more likely to bunt in past seasons. You can then refine your replay, going from forcing yourself to bunt a certain number of times per game to ensuring that your strategy matches real-life strategy as closely as possible.