Condolences to those still recovering from staying up to watch that slog of a game 4 last night/this morning. There are plenty of things to talk about from last night’s game, but the one I wanted to focus on is a strategic decision made by Craig Counsell all the way back in the 7th inning.
Manny Piña led off the inning with a double, which brought up Orlando Arcia. Rather than having Arcia bunt Piña over to third, Counsell let him swing away. Arcia flew out, and the Brewers ended up stranding Piña on second.
Later in the broadcast, we were unfortunate enough to witness Alex Rodriguez’s attempt at analysis, which he of course bungled by both misspeaking and misrepresenting the actual situation (but he did have visual aides!). His argument was that it was clear that letting Arcia swing away was the wrong decision. Let’s take a look at whether that is the case.
A good way to do this is by the use of a run expectancy matrix. The most complete one I’ve been able to find is this one from tangotiger, so that’s what I’ll be using to come up with these figures.
The reason that most sabermetrically inclined baseball fans are anti-bunting is because it generally lowers the amount of runs you’re expected to score in a given inning. On average, a runner on second base with nobody out results in 1.100 runs per inning, while a runner on third base with one out scores 0.95 runs per game. In other words, sac bunting damages your chance at a multi-run inning, as it gives up one of the three outs with which you can work in an inning.
However, it’s fair to point out that at that point in the game, the Brewers did not need a big inning, and that one run could very well have been enough to win the game. So we should also consider the odds of scoring at least one run in these same situations. So, a runner on second with nobody out yields a 61.4% chance at scoring, while a runner on third with one out yields a 66% chance of scoring at least one run.
So, that’s a pretty clear advantage to bunting in that situation, right?
Simply plugging in those figures assumes that a sac bunt is 100% successful, which it of course isn’t. A batter could fail to get a bunt down in two pitches and then have to swing away in a disadvantageous count. He could strike out while attempting to bunt with two strikes. He could pop the bunt up. Or worst yet, it could result in the runner being thrown out at third. So that definitely makes things murkier.
Another way to analyze this is with a win expectancy matrix. The frame of reference for this calculation is the visiting team up to bat in the top of the 7th inning of a tie game.
A runner on second with nobody out results in a 60.17% chance of victory for the visiting team. Meanwhile, a runner on third with one out results in a 60.61% chance of victory. Still a slight edge to having the runner on third with one out, but not nearly as large as the run expectancy chart might indicate.
The following aspects were working against the Brewers pulling off a successful sac bunt:
- They don’t bunt with position players, almost ever. They ranked 14th in the NL in sacrifice hits in 2018.
- Arcia, despite being a light hitter, has one sac bunt this season and three in his entire career. He doesn’t do it often at all.
- Piña is slow and therefore more likely to be out at third than a faster runner would be.
- This one should be a much smaller factor, but Arcia has been (one of the only Brewers) swinging a hot bat recently.
Given these factors and the fact that a successful sac bunt would have only increased their chance of winning by just 0.44%, the reward simply doesn’t outweigh the risk. It was a close call, but Counsell made the right decision. It just didn’t work out.