It’s a play as old as the game itself, and maybe that’s because the DH didn’t always exist. It’s the bunt, a time-honored tradition that is being phased out of the game more and more every year.
Fangraphs has sacrifice bunt totals dating back to 1895, and according to their database, the 925 sacrifice bunts laid down by teams in 2017 was the second-lowest figure in MLB history. Only the 1900 season (806 bunts) featured fewer, and there were only eight teams back then.
So yeah, the bunt is dying a slow death.
At a glance, the Twins were in the middle of the pack as far as bunts were concerned, checking in 17th among 30 MLB teams with 26 sacrifices in 2017. That’s a deceptive number, though; flip the dial to only AL teams, and only two teams — the White Sox and Rangers — bunted more often than the Twins.
That’s more problematic. Subtracting NL teams — yay, pitchers hitting! — from that figure shows how much more the Twins were devoted to bunting than their junior circuit contemporaries.
To frame up how different the game is bunts-wise, consider this:
- Twins bunts as a percentage of MLB on the whole: 2.8 percent (one team = 3.3 percent of MLB)
- Twins bunts as a percentage of AL on the whole: 9.6 percent (one team = 6.7 percent of AL)
As you can see, bunting is severely shifted toward the NL, and the Twins are well above the average mark of their AL contemporaries.
The White Sox offense was absolutely dreadful in 2017, scoring 109 fewer runs than the Twins while the team lost more games than every team in the AL but the Detroit Tigers. The Rangers were in the thick of the race for much of the season despite Adrian Beltre missing time due to injury and a patchwork rotation, and scored just 16 fewer runs than the Twins while winning 78 games to Minnesota’s 85.
So maybe everything said in this space will also apply to the Rangers — we’ll see.
Here’s one thing that stands out, and it is glaring: no team bunted more than the Twins did with their No. 3 hitters. In fact, according to Baseball Reference, the Twins bunted five times with their third hitter. The rest of the league combined bunted six times with their No. 3 guy — and no team did so more than once.
And according to Parker Hageman of Twins Daily, none of those bunts led to runs. Blech.
So what’s the big deal? It’s just giving up an out to move up a base. Doesn’t it make it more likely the team scores a run? Doesn’t it lead to more scoring? Aren’t those basically asking the same question?
For this, we’ll consult the run expectancy matrix:
The latest edition of the matrix splits into four run environments as listed by the dates at the top. The 2010-15 edition is the most recent, and frankly there isn’t a ton of change from era to era. If anything, the value of giving up an out for a base has gone down with how much the ball flew out of ballparks in the 2017 season, but for now we’ll treat that as an outlier and assume the run environment has played pretty close to that six-year span.
If the table is a bit daunting to read, we’ll explain it a bit. We’ll use the same hypothetical for each situation — runner on first, no one out. That seems to be the most obvious bunting situation, anyway.
What the matrix says is that the average number of runs that scored in an inning with a runner on first and zero outs is 0.859. With a runner on second base and one out — the same as a sac bunt — that drops to 0.664. In short, the number of runs that score from that base/out state to the end of the inning drops — making the bunt a poor proposition. To over-explain, it lowers the number of runs a team scores in those innings, on average.
The second table says that with a runner on first and nobody out, the chances of scoring a run in that inning are 41.6 percent. Bunting the runner to second drops the scoring percentage to 39.7 percent.
In short, again, bunting hurts run scoring. It makes it less likely a run scores.
Now, this isn’t to say there aren’t spots where a successful bunt can be helpful. Now keep that in mind — a successful bunt. A lot of times the war that rages between traditional and contemporary baseball thinkers is with the bunt viewed as a 100 percent play. Bunts can be defended, hitters can be bad at bunts and myriad other issues can be at play which can hurt the overall effectiveness of a bunt — none of which is factored into these numbers.
One spot where a bunt can be helpful is when there’s a runner on second and none out. In that base/out state, the team scores an average of 1.1 runs before the bunt and 0.95 after. Wait, isn’t that worse? Well, yes and no. The chance a run scores in the inning goes up from 61.4 percent to 66 percent. In short, it’s a trade-off of a base for 4.6 percent of a chance to score. In layman’s terms, it means it’s defensible if only one run is needed.
But that’s where issues can come into play as we step outside of the vacuum.
When is only one run needed? For the 2017 Twins, it’s a complicated question. Any time a bunt is called for it’s a maligned decision because it’s asking for the team to give up an out to a pitcher who didn’t earn it — and right after someone did something to him like walk or get a hit. When a pitcher is wild, giving him an out is inadvisable. Then too, wouldn’t giving an out to a pitcher who just gave up a double seem strange as well?
But that’s not where Pandora’s box stops with this situation, either. A successful bunt moves runners up, but it also opens up bases where a team can put on a particularly hot or overall good hitter to face his inferior or possibly slow-footed lineup mate behind him. In today’s age of specialized bullpens, a lot of teams have guys who are designated groundball inducers, who can come in and wipe the bases clean with one grounder. Trevor Hildenberger is that sort of pitcher for the Twins, and when a team has already given up the first out of the inning to move a guy up, they have no more to give with a double play induced ending the inning — and thus the threat.
And again, to hone back in on the 2017 Twins, when exactly is just one run necessary? The Twins bunted like that was a frequent situation this past season, but frankly, unless a team is trying to get a walk-off win, it’s worth considering the weight of getting just one run against what could happen if the offense is just left to its own vices. Again, that’s even apter when considering how many home runs were hit in MLB in 2017.
Bunting can be a way of scratching across runs for a weak offense — again, it’s not efficient as noted by the matrix — but that doesn’t apply to the 2017 Twins, either. Even with virtually nothing from Miguel Sano after the All-Star break, the Twins were seventh in MLB in runs scored, ninth in wOBA (weighted on-base average, via Fangraphs) and tied for fifth in wRC+ (weighted runs created plus, also Fangraphs).
No matter how the apple is sliced, this was a well above-average offense — one that need not be reduced to run-scoring ploys. It alone is self-sufficient, and scarily enough, has room to improve with the return of Sano and potential season-long improvements of Jorge Polanco and Max Kepler.
The other thing is that playing for just one run at a time isn’t just based on a team’s offense. If a game is tied 4-4 in the seventh and a team bunts to try scratch across a run, they’re also indicating that they think their pitching staff — mostly bullpen — is good enough to hold that one-run lead for the final two or three innings, depending on if it’s a home or road game.
That might be true of the Twins bullpen moving forward, but that’s not really true of a group that posted a collective 4.40 ERA (20th in MLB), 7.7 strikeouts per nine innings (29th) and a collective fWAR of 2.2 (21st).
Playing for multiple runs with a good offense is just good for business; playing for multiple runs with a leaky bullpen just fortifies that stance.