Which bullpen has the edge between Indians and Red Sox?

The Cleveland Indians and Boston Red Sox each had a good bullpen during the regular season, and the two groups posted pretty similar numbers. Tribe relievers had a collective ERA of 3.45 and FIP of 3.66, with 9.09 strikeouts per nine inning, 3.39 walks per nine innings, and 0.94 home runs per nine innings, while the BoSox group had a 3.56 ERA, 3.67 FIP, 9.70 K/9, 3.69 BB/9, and 0.90 HR/9. Both teams play their home games at ballparks that favor hitters, and adjusting for park factors doesn't do anything to separate the two, as their ERA- and FIP- are each only one point apart. Taken as a whole, each of these bullpens was among the top five in the American League, with little to separate them from the other groups in that tier.

Each bullpen was also especially good after the All-Star break, with Boston posting an AL-best ERA- of 71 (the Indians were second) and Cleveland posting an AL-best FIP- of 79 (the Red Sox were fourth). First-half vs. second splits for individual relief pitchers feature sample sizes too small to be worth much analysis, so other than in the cases where an injury is a factor, I wouldn't worry about them, except to note that relievers who appear in a lot of games will sometimes wear out. The reason each team's second-half improvement is notable isn't about guys pitching better, it's about each team adding a new arm to the bullpen. In July the Indians acquired Andrew Miller from New York, and the Red Sox acquired Brad Ziegler from Arizona. In September the Red Sox got a healthy and effective Koji Uehara back. Those guys were all excellent down the stretch.

That brings me to what we really ought to focus on when examining bullpen strength in the postseason: not the group's overall performance, but the individual quality of each team's best and most depended-upon relievers. There are more days off, and each game is of supreme importance, so teams can and usually do rely on their top relievers for almost every important relief inning. Not counting a couple of long-relief appearances in lopsided games, last year the Royals' top four relievers pitched 82.3% of the team's relief innings.

There's always a chance of a long extra-inning game that leads each manager to go beyond those top guys, but the top guys matter far, far more than the bottom guys, so let's look at each team's top relievers

Boston

*Shutdowns and meltdowns are a metric found at FanGraphs, and which use WPA (win probability added). In short, if a relief appearance improves a team's chances of winning by 6% or more, it's a shutdown. If an appearances hurts a team's chances of winning by 6% or more, it's a meltdown. You can read more about them here.

Kimbrel was good this season, but his walk rate was significantly higher than it had been in the past, and he's been giving up home runs with greater frequency during the last couple years. Uehara was excellent down the stretch. It's a small sample, but after returning from the DL in early September, he made eleven appearances, each exactly an inning long, and didn't allow a single run. He struck out 12 while walking only two.

Cleveland

Miller is the only left-handed pitcher the Indians have on their ALDS roster, and David Ortiz is by the far Boston's biggest left-handed hitting threat (with Jackie Bradley Jr. a distant second, but also dangerous). We should expect to see Miller brought in for an inning in which Ortiz will bat every game this series (or at least every game that's remotely close). Terry Francona has shown he's willing to use Miller in whatever situation he think is most important, even if it's only the sixth inning, and also that he's willing to lean on Miller for more than three outs. Miller might be the best reliever in baseball, and is likely to be a key part of any Tribe success.

Shaw has pitched in 299 regular season games during the last four seasons, the most of any pitcher in MLB. Early this season it seemed like his workload might be catching up with him, but after couple terrible outings early on, he recovered to post another strong season. Otero is something of an odd bird, because few relievers succeed with a relatively modest strikeout rate, but if he can avoid walks and home runs the way he has, the Tribe's strong infield defense allows him to succeed with a lot of balls put into play.

Boston's lineup poses a massive challenge to any pitching staff, but especially to one without two of its three best starters. The Indians are likely to need to lean on their bullpen a lot in this series. Fortunately, their top four relievers have been up to the task this season, and if the starters and offense can keep games close into the later innings, the team will have a good chance of holding on or keeping things close enough for some late-inning heroics by a team with an MLB-best 11 walk-off victories this season.