Finesse in Baseball
As a tennis player, I appreciate the use of finesse to win points. I don’t think I hit the ball particular hard, but I think I am relatively good at placing my shots. So I appreciate a player like Roger Federer who I believe is remarkably good in using finesse in being successful.
In baseball, a bunt is a finesse play. I think a bunt gets a bad rap since it is associated with a sacrifice hit that seems not be very popular in baseball. So I think it would be interesting to look at bunts and see how the use of bunts and sacrifice bunts has changed over the past 50+ seasons.
For a particular Retrosheet play-by-play data frame, I focus on the following variables:
- BUNT_FL – a flag variable indicating if a bunt occurred
- SH_FL — a flag indicating if there was a sacrifice hit
- SH_Bunt — this indicates if a bunt was hit and it was a successful sacrifice
- H_Bunt — this is bunt that is a hit
- BAT_EVENT_FL — this indicates if a batting event occurred — the count of these will be the number of plate appearances
For each of the seasons 1960 through 2015, I find the sum of these events and use these sums to compute specific rates.
Demise of the Sacrifice Bunt
Most of us are aware that the use of the sacrifice bunt is decreasing. To confirm this, I plot the fraction of PA’s that are sacrifice bunts for the seasons 1960 through 2015. I added a red vertical line at the year 1973 — this is the year that the American League started using the designated hitter.
As one expects, the rate of sacrifice bunts has been decreasing — in fact the lowest rate is 2015. It is interesting that the use of the DH did not result in an immediate drop in SH’s — the drop started around 1980. Then for some reason, the rate of SH’s increased until the middle 1990’s and then it dropped, hit a plateau between 2000-2010, and then started dropping again in recent seasons.
Do We See the Same Pattern in Bunt Rates?
What if we instead focus on the fraction of PA’s that are bunts? In the following graph, we see a different pattern.
- From 1960 through 1980, the bunt rates were pretty constant.
- For some reason, the bunt rate suddenly dropped in the early 80’s, and then it returned to a normal level in 1990.
- Then it dropped steadily from 1990 through 2000, increased again, and then has been dropping in recent seasons.
Purpose of Bunts?
There seem to be two primary purposes of bunts — one purpose is to sacrifice to advance the runner or runners, and a second is to get an infield hit. This motivates looking at the fraction of bunts that are (successful) in getting a SH.
In the early seasons, the sacrifice rate (among all bunts) was in the low 60% range and increased to over 70% in the early 1980’s. Then for some reason, the rate dropped sharply to the low 50’s, and the rate has been close to 50% in recent seasons. There is an interesting outlier in the 1980’s where the rate was over 80%. I don’t see any obvious DH effect since the sacrifice bunt fraction stayed pretty constant in the 1970’s — or maybe there is a DH effect, but it took awhile for baseball to adjust.
Using the Bunt to Get a Hit
One characteristic of “small ball” to use a bunt to get a hit. In the following graph, I plot the fraction of bunts that are hits for all seasons.
We see a new type of pattern. In the early seasons, we see an initial drop followed by a gradual increase until the early 80’s. For some reason, there were two low rates in 1982 and 1983. In recent years the fraction of hits has been relatively constant, but the pattern is steadily increasing the last 10 seasons.
Perhaps I brought up new questions than I answered by constructing these graphs. The use of bunts and sacrifices may be related to the scoring — perhaps when scoring is down, players might be more inclined to bunt to get base hits. I wonder about the effect of shifting. When a team does a full infield shift, it seems that they are daring the left-handed hitter to bunt to the left side of the infield to get a hit. If this is happening, then one would think we’d see an increase in bunts for hits. In a future post, I’ll look further at bunting for a specific season. Then I can explore the variability of bunting across teams and find the situations where bunts are more or less common.