In order to understand why Major League Baseball has announced there will be a three-hitter minimum for each pitcher (barring injury or the end of an inning) coming to the game in 2020, look no further than a White Sox/Angels game from last July 23.
To kick off the bottom of the eighth inning in Anaheim, Chicago righty Juan Minaya allowed Albert Pujols to single to left. Out came Chicago manager Rick Renteria, signaling to the bullpen for his fourth pitcher of the night. He'd soon wear a path between the dugout and the mound.
Here's how the eighth inning ended up playing out, in full.
1) Minaya allows Pujols to single.
Renteria signals for a reliever. Two minutes and five seconds pass before the next pitch.
2) Jace Fry enters to relieve Minaya. He throws six pitches, striking out Shohei Ohtani.
Renteria signals for a reliever. Three minutes and 27 seconds pass before the next pitch.
3) Jeanmar Gomez enters to relieve Fry. He throws two pitches, getting Ian Kinsler to pop out.
Renteria signals for a reliever. Four minutes and 10 seconds pass before the next pitch, in part because the Angels had sent lefty Luis Valbuena to hit for Martin Maldonado, then replaced Valbuena with righty Jefry Marte after the White Sox stalled to buy time to change pitchers.
4) Luis Avilan enters to relieve Gomez. He throws eight pitches, striking out Marte to end the inning and his night, since Joakim Soria enters for the ninth inning.
"It's empty-the-bench Monday for the Angels and White Sox in the series opener," said White Sox announcer Jason Benetti as Avilan entered. All told, the half-inning took 17 minutes and 24 seconds from Minaya's first pitch to Avilan's final one. That breaks down into "7:02 of baseball time and 10:22 of not-baseball time," with the latter including 6:14 for mid-inning commercials, 2:13 waiting for relievers to finish warmups after returning from break, and 1:55 for Renteria to walk to the mound three times.
(You can see the full breakdown of the inning here.)
This is an extreme example, we'll grant. This doesn't happen in every game every night. But this is the point of all this, sequences like this. Mid-inning pitching changes are a plague upon pace, and the game. They stop the action, and add none in return. The fewer of them we see, the better.
More pitchers, more mid-inning breaks
In 2018, there were 799 different pitchers to appear in a game, a new all-time record. This is a record that's been broken each year since 2013, unsurprisingly. If we go back to 1998, the first year of the 30-team era, there were 557 pitchers, appearing in 4,864 games. In two decades, in a nearly identical number of games, the number of pitchers per game has shot up by 2.6, from 6.1 to 8.7.
That's not going to change, for the most part. Starters aren't going to suddenly throw 300 innings again. It's not happening. But the increase in pitchers, or more accurately relievers, has also increased the amount of pitching changes.
If we just look at the number of relief appearances that have lasted a maximum of two batters, we can see that's massively changed as well. Before World War II, there were usually fewer than 200 such appearances across the entire sport. As recently as 2004, we were below 2,000 total appearances. Now, we're routinely in the 2,300 to 2,500 range.
On a per-game basis, it's not quite as stark, obviously, because the number of teams and games have both increased over time, and by that view, this is still happening less than once per game. It's also true that by percentage of relief appearances, those of two batters or less has actually gone down in the last few years, but that's also a function of so many, many more overall relief appearances. As a raw total, there are more of these than ever.
If there's an argument that this rule changes the game too much, that might not be wrong, but the point here is that the game has changed, considerably. Whatever you think "the right point in baseball history was," the game in 2019 doesn't look like that. Years ago, there weren't games with eight relievers. There weren't pitchers in for just a batter or two. If anything, this might make baseball look a little more like it used to. Change, good or bad, always happens.
What we showed above was simply "relief appearances of two batters or less," because that's a good way to show this impact over time, but that's not exactly the rule. As stated in the rules, it will actually be a requirement that a pitcher face "either a minimum of three batters or to the end of a half-inning."
So, how much impact will this actually have? Actually, perhaps not quite as much as you'd think.
How much will this actually change things?
We need to split this up into "relief appearances of zero, one, or two batters where the inning is ended," thus freeing the reliever from the requirement to stay in, and "where the inning is not ended," because that's now important. If we look back at the last 20 seasons, we can see that the number of relief appearances that don't stretch to three batters has increased markedly ... but the number of appearances where a pitcher did that and didn't get to the end of the inning hasn't.
Basically, we're looking at about 800 or so of these soon-to-be-banned relief appearances each year, plus a small impact to the length of "opener" appearances, because we're looking just at relievers for this data. While that hasn't changed appreciably over time, it's still notable. There are 26 weeks in a Major League Baseball season, and we saw 779 of these appearances in 2018. That's about 28 times per week, or roughly one per team per week. It's not a lot. It's not nothing.
As Matt Eddy of Baseball America showed, we've already seen a slight increase in batters faced per relief appearance, likely in part because if starters are throwing fewer innings, you need some longer relievers to pick up the slack.
The outcome here might be to harm LOOGY types -- that's "Left-Handed One-Out Guys" -- like Andrew Chafin or Jerry Blevins, but it probably increases the value of those relievers who can go multiple batters without large platoon splits, like Josh Hader or Andrew Miller. If this shifts value from specialists to better overall pitchers, all the better.
It doesn't end strategy, it changes it
One common argument against this idea is that it limits strategy, because managers will no longer go through the chess match we described above in the White Sox/Angels game. Whether or not that's a good or bad thing is up to your own interpretation, but it might also not be true.
Consider this hypothetical situation: There are two outs and men on. The next three batters swing lefty/righty/righty. The manager has two interesting options...
A) Bring in his ace lefty-killer to try to end the inning. In this scenario, he's trying to end the inning immediately, knowing that his pitcher has a higher chance of getting out the first hitter, but may be weaker against the next two.
B) Bring in a better overall pitcher. In this scenario, he's putting out a pitcher with a worse chance of getting out the first hitter, but a better chance of getting out the next two that he'd be forced to face.
Obviously, pinch-hitting possibilities factor into this too, further affecting the manager's decision. Strategy won't go away. It will just be different. Lefties won't go away either, because you're not going to not bring in a lefty to face a Cubs grouping of Kyle Schwarber, Kris Bryant, and Anthony Rizzo, for example. It'll just be different lefties, ones who can do more than one thing.
We can look at which teams will be the most affected, too.
There's an interestingly wide spread there, from 52 from the Indians -- thanks in large part to Oliver Perez, who was pulled from a soon-to-be-outlawed appearance 19 times -- all the way down to the Marlins, who did it just seven times.
Now, it's fair to say there's a different way to get to this. You could argue for a penalty, say a ball added to the count, for a mid-inning pitching change, or merely a limit on those changes entirely. Those ideas work too. But no one's arguing that it's fun to sit through a stream of changes where the manager slowly walks to the mound, the telecast goes to break, and minutes pass without a pitch. This won't fix all problems. It will fix this problem. That alone makes it worthwhile.
Mike Petriello is an analyst for MLB.com and the host of the Statcast podcast.