The Three Batter Minimum Rule’s Biggest Dilemma
The three-batter minimum rule is coming to a regular season game near you in just a couple weeks. The stated desire of the rule is to reduce those time-consuming and action-relieving breaks late in games as a parade of relievers come in to get just a couple of outs. (The rule, for those who need a refresher, requires pitchers to either face a minimum of three batters in an appearance or pitch to the end of a half-inning, with some exceptions allowed for injury and illness.) If pitchers are forced to stay in games, then we’ll end up with fewer pitching changes and fewer breaks. That’s the idea, anyway. Ben Clemens took a look back in December and found that the number of times the rule would have actually come into effect in 2019 was actually pretty minimal. But that doesn’t mean there won’t be an effect on strategy.
Over at The Athletic, Jayson Stark talked with multiple managers to get a handle on how much they’ll have to plan, and to figure out how the new rule could affect deployments when it comes to lineups, pinch hitters, and when to use relievers. It seems the rule will invite intentional walks and, as Buster Olney had previously mentioned, mid-batter pitching changes as pitching the last ball to a batter in a walk gets a pitcher credit for facing a batter. In some cases, this could even result in the return of pitchers using four actual pitches to intentionally walk a batter in order to set up the next reliever to come in before the intentional walk is complete.
While the resulting strategy will be much-discussed and analyzed in the context of a number of game-states, there is one situation that interests me the most: two outs, late in the game, with runners on base and a good lefty hitter coming to the plate.
Under the old rules, we might see a lefty come in to try to finish the inning; failing that, a righty could follow for the next batter, and try to finish things off. Now, if the lefty comes in and does his job against a lefty batter, the minimum is successfully skirted and the pitcher doesn’t have to start the next inning. However, if the lefty fails to retire the batter, he must stay in to face another hitter, presumably a righty in an even more difficult circumstance.
To test what a manager might decide to do, I set up a two-batter situation. First, we’ll assume the home team has a one-run lead in the top of the eighth inning. We’ll further assume that there are runners on first and second base, and that the next two batters are very good hitters, with a lefty up followed by a righty. While I don’t necessarily have a single matchup in mind and concede that these might not be possible depending on how the teams set up their orders, think something like Cody Bellinger followed by Mookie Betts, or Anthony Rizzo batting in front of Kris Bryant, or Freddie Freeman and Ronald Acuña Jr. Many good teams have pairs that come pretty close to those three in their quality, and I wanted to set up a situation where neither player would be pulled for a pinch hitter.
To set a talent level, I took the top 12 lefty batters overall and the same number of righty batters to see how each set operated versus platoons last season. Here are the general splits:
AVG | OBP | SLG | wRC+ | |
---|---|---|---|---|
Top L v RHP | .302 | .402 | .587 | 155 |
Top L v LHP | .276 | .359 | .511 | 127 |
Top R v LHP | .302 | .411 | .613 | 163 |
Top R v RHP | .295 | .386 | .581 | 151 |
Both sides perform better with the platoon advantage, but the difference is much bigger for lefties than it is for righties. While this effect isn’t present when taken across all players, lesser hitters (particularly lefties) are more likely to be deployed in a more strategic fashion to avoid mismatches, while top hitting lefties are good enough to always stay in the lineup. And the same effect is seen to a lesser degree when looking at decent or better relief pitchers. I made small changes to the numbers above to factor in the higher quality of pitching seen in a potentially high-leverage situation, but the platoon effect wasn’t huge, i.e. good righty relievers generally do pretty well against lefty bats and good lefty relievers do pretty well versus righty bats, though not quite as well as they do with the platoon advantage.
That places most of the difference in the platoon advantage on the batter. Managers in this situation have three choices. The first is to bring in a lefty and if the pitcher doesn’t get the batter out, risk a lefty-righty matchup. The second is to keep the righty in (or bring in a different righty) to face the lefty; if the plate appearance goes wrong, the favorable righty-righty matchup follows. The third is to intentionally walk the lefty and go straight to a righty-righty matchup.
Before we get to the intentional walk, let’s look at the first two scenarios. The results below show how many runs came across the board for the next two batters, and if the inning was over or still going. The outcomes are sorted based on what is most favorable for the pitcher:
Runs Allowed | Inning Status | Lefty RP Outcomes | Righty RP Outcomes | Difference |
---|---|---|---|---|
0 | Over | 71.96% | 70.33% | 1.63% |
1 | Over | 8.38% | 7.55% | 0.83% |
1 | Still Going | 4.15% | 3.98% | 0.17% |
2 | Over | 2.88% | 2.87% | 0.01% |
2 | Still Going | 4.43% | 4.69% | -0.26% |
3 | Over | 2.47% | 4.05% | -1.58% |
3 | Still Going | 3.64% | 4.17% | -0.53% |
4 | Still Going | 2.09% | 2.36% | -0.27% |
No matter who the pitcher is in these situations, the outcomes mean either a very good chance of a win or a very good chance of a loss. In a majority of the situations above, the home team takes an 87% win expectancy to the bottom of the eighth inning, with another 8% or so going to the bottom of the inning tied and a better than average shot at winning. The number of scenarios in which we get past the first two batters and the third batter can even make a difference occurs maybe one out of 12 times regardless of whether a righty or lefty is used. We can see the two most favorable outcomes above happen when a lefty is brought in. Here’s how that translates to Win Expectancy, including the intentional walk scenario:
Decision | Win Expectancy |
---|---|
Bring in Lefty | 72.60% |
Go with Righty | 71.10% |
Walk Lefty with Righty | 65.40% |
While it’s seemingly risky with a righty on deck, the best move is to bring in a lefty for a lefty-on-lefty matchup. Getting the immediate out is the best move when possible, and counteracts the potential advantage of a righty-righty matchup to follow. And whatever you do, don’t intentionally walk a batter when the following hitter is good. Another baserunner means more potential damage when your pitcher is facing a hitter who is likely to inflict said damage on your team.
There are a number of different but similar situations that could affect a team’s chances here, but they generally aren’t going to move far away from favoring getting the immediate out. If a team has a much worse right-handed hitter coming up after a strong lefty, all the more reason to try and get the lefty out with the best matchup possible; if it goes awry, then a weaker batter and a still decent matchup follows, even without the platoon advantage. If the lefty hitter is much weaker than the righty behind him, and is likely to be pinch hit for if a lefty comes in, then sticking with the righty makes sense, but if the lefty is a good enough hitter not to be pinch hit for when a left-handed reliever comes in, the left-handed reliever likely provides the greatest chances of success.
Just to provide another example, here’s the same situation, but with just a runner on second base:
Decision | Win Expectancy |
---|---|
Bring in Lefty | 76.3% |
Go with Righty | 75.1% |
Walk Lefty with Righty | 73.0% |
The difference isn’t as great, but intentionally walking the batter with a good hitter behind him in the order is a bad idea. Even if the hitter behind the big left-handed hitter was merely average, going to the lefty is still a better play by more than one percentage point of win expectancy, with the intentional walk and sticking with the righty only though the latter two options are almost indistinguishable.
Going to a lefty knowing they might have to face a righty in an important situation feels like a risky move, but as is the case in most situations, getting the out when you can is the better play.
Craig Edwards can be found on twitter @craigjedwards.
This is really interesting work. Nice job! I wonder what it would look like in, say, the 6th or 7th when a potential reliever may have more pronounced platoon splits? Does this basically only apply to the great relievers?