So that was a pretty enjoyable postseason. A terrific World Series, with one of the best Game Sevens of all-time, wrapped up a month of high-quality baseball, with only a few duds mixed in here and there. As usual, it was a low-scoring month, with cold weather and elite pitchers serving to make offense scarce, but that just makes for more tense, high-leverage innings.
Of course, there was one notable change this year, particularly emphasized because of Cleveland’s run to the championship. More than ever before, managers were willing to use their bullpens without regard for role or inning, with Andrew Miller serving as the platonic ideal of a relief ace. It wasn’t just Miller, though; Cody Allen entered in the middle innings a few times, while Kenley Jansen and Aroldis Chapman both entered in the seventh inning in several outings.
Some of this change was the inevitable rationalization of Major League organizations, as the concept of strict relief usage has never really been the optimal way to run a bullpen in the postseason. That was an idea in need of challenging, and it was only a matter of time before the incentives to win overcame the notion that relievers could only be used in the way they were deployed in the regular season.
But the early appearances from the game’s best relievers also had another cause; starting pitchers were removed earlier than ever, with team’s utilizing the times-through-the-order information to aggressively remove pitchers even when they’re throwing well. This was most obvious in the final game of the year, with Kyle Hendricks getting lifted after throwing just 63 pitches to 19 batters despite a comfortable four-run lead. Hendricks was one of the best pitchers in baseball this year, was cruising through Cleveland’s line-up, and the game wasn’t in any real jeopardy, but Joe Maddon went and got him anyway, not wanting to experience the third-time-through rally that is seen so often in the regular season.
Rob Arthur did a nice piece for FiveThirtyEight about this halfway through the postseason, showing the declining number of innings starters were throwing this year compared to prior years. At the point of publication, he showed that starting pitchers were averaging about 5.2 — that’s in decimal form, not 5 2/3 innings — innings per start, but now that the World Series has been played, the final total for 2016 is 5.1 innings per start, the lowest mark in baseball history.
Or, if we want to look at it another way, relievers threw 43% of all the postseason innings this year. That’s up from 39% a year ago, and over the last five years, it had held pretty steady between 35-40%, right in the same range as the 63%/37% split we see during the regular season. It certainly feels like this year is the start of a shift, and teams are going to continue to push their bullpen usage in October, especially given how well it worked this year.
But that brings up a bit of a philosophical question; how far away from regular season baseball are we comfortable letting postseason baseball get? It’s already very different in some ways, most notably the number of starting pitchers a team needs to get through, based on the huge increase in off-days relative to what we see from April through September. But with the league potentially moving towards a 50/50 split in workloads between starters and relievers in the postseason, is there a point at which we’ll wonder if October baseball is just too different from the one teams have to play in order to get there?
Some change between the regular season and the postseason is inevitable. When managing for 162 games, the incentive to win any single game is diminished, so there just isn’t the reward to justify the risks of extreme reliever usage that we just saw. In the postseason, every win is vital, and so the difference in urgency is always going to drive a different type of decision making. Toss in the extra days off, and the recipe is there for bullpen-heavy pitcher usage.
But from a roster construction standpoint, it’s a little bit awkward to be incentivizing very different players in October. While elite relievers are obviously extremely valuable in the playoffs, their value remains muted in a regular season where they can only throw 60-70 innings out of the 1,450 a team needs to get through the season. The Yankees, for instance, were built perfectly for October, and might have run over the rest of baseball had they reached the postseason with Andrew Miller, Dellin Betances, and Aroldis Chapman. But even with three high-end bullpen arms, the Yankees were just a .500 team at the trade deadline, which is why Chapman and Miller ended up pitching for Chicago and Cleveland in the World Series.
Building a team for the 162 game haul is quite different than building a team for the playoffs, so it seems like at this point, the general plan is to build a good regular season club and then try to acquire relief help at the deadline. But perhaps the recognition of that plan is what drove reliever prices up so high last winter; if you’re contending, well, great, you already have good bullpen arms, and if you aren’t, you now have the asset that everyone else is trying to trade for in July.
And with few good starters and a trio of high-end closers available this winter, it seems likely that reliever prices are only going up against. Jansen and Chapman are probably going to threaten — and maybe break — the $100 million mark, which no reliever has gotten anywhere near in previous years, and Mark Melancon is going to get paid a pretty penny too. But the teams that make big investments in these guys are going to have to hope the rest of their rosters are pretty well set, because the payoff for having these guys around won’t come if the team doesn’t make the postseason.
But the only real way to push innings from the bullpen back towards starting pitchers would be to have fewer off days in October. And that would require changing the postseason schedule pretty dramatically. Going from a 2-3-2 format to something like a 3-4 format would eliminate one off day from each series, and taking away that second rest day would make it more difficult for teams to extend their bullpens in the way they’re being used now.
Going to one off-day per series would have some other benefits, either allowing the World Series to end sooner — and avoiding the issues that come with unpredictable November weather — or by cutting down the schedule enough to expand to the best-of-three Wild Card format that many have advocated for. After all, when we’re talking about the difference between the regular season and the postseason, the Wild Card play-in games are the least regular-season like thing baseball has going right now.
But they’re also a lot of fun. I know they’re not particularly good at determining which team had the better season, but from a viewer standpoint, winner-take-all games are amazing theatre. And baseball, at the end of the day, is an entertainment product. And while it isn’t anything like the regular season, postseason baseball is still a wildly entertaining way to end the year.