October Bullpens Now Carry Almost Half the Innings, and Rest Decides How Well

Through two games of the NLCS, Milwaukee's relievers have done a remarkable share of the work. In Game 1, four of them finished a 2-1 win after Jacob Misiorowski was chased in the fifth inning. In Game 2, four relievers retired all 15 Dodgers they faced in a 7-4 win that William Contreras tied twice and then won with his bat.

That is not unusual by modern standards. What has changed over the last decade is how much of a postseason series belongs to the bullpen, and the research on why that happened also identifies where it goes wrong.

The shift in numbers

A 2023 study in SABR's Baseball Research Journal found that starters threw 63 to 64 percent of innings in League Championship Series and World Series games from 1996 through 2013. From 2014 through 2022, that share fell to 54 percent. Relievers now handle close to half the innings in a seven game series.

The main reason is the times through the order penalty. Hitters improve each time they see the same pitcher in a game, by roughly 10 to 15 points of wOBA per trip through the lineup. Managers who know that pull starters earlier, especially before a third look at the top of the order.

The cost nobody counted

The study's point is that the penalty has a counterpart. In a seven game series, the same relievers face the same lineup again and again, and they get tired.

The author sorted relievers by recent workload. Those who had pitched on consecutive days, or faced five or more batters in the previous two days, allowed a .319 wOBA, which was .076 higher than in their first appearance of the series. Rested relievers allowed only .017 more than in their first outing, a gap too small to be statistically meaningful. Relievers with two or more days off showed no significant decline at all.

Relievers also lost effectiveness the more often a lineup saw them. By a third appearance in the same series, the increase in wOBA was larger than the penalty a starter pays the third time through. The author estimated the starter's cost at about .030 wOBA, well below the fatigue cost for overworked relievers.

What that means for this series

Milwaukee's bullpen has been excellent, and an off day on Tuesday between Game 2 and Game 3 helps. But a single day of rest is the category the study flags. Relievers who face five or more batters and come back after one day off showed one of the largest drops in the data.

That does not mean the Brewers bullpen is about to fade. It means the strength on display so far comes with a schedule attached. If the same arms are asked to cover four or five innings again on Wednesday and then again later in the week, the evidence says the results get worse.

The honest caveat

The study is one author's analysis of postseason samples, which are small by nature, and the recommendation it makes is a judgment rather than a rule. It argues against pulling a starter automatically at the third time through when he is pitching well, not against bullpens in general.

Still, the logic is hard to dismiss. October rewards teams with deep relief, and it punishes teams that use the same five relievers every night. The difference shows up late in a series, not early.