December 17, 2015
The Rise of the LiRPA
In 2015, 137 pitchers threw at least 48 2/3 innings in relief. Of those 137 pitchers, Fernando Rodney was 134th in terms of RE24[i] , pitching roughly 10 runs worse than the average reliever in baseball. Despite being objectively awful, Rodney had the 11th highest inLI—the leverage of the situation when he entered—of those same 137 relievers. Rodney notched 16 saves as the Mariners closer before ending up in Chicago, where he was surprisingly dominant for a handful of innings at the end of the year. But in Seattle, Rodney’s track record and closer job title garnered him plenty of high-leverage innings despite his being one of the worst possible options to pitch them.
Luckily, Rodney didn’t lead the team in inLI. That honor goes to Carson Smith, whose 2.11 inLI was actually the highest in baseball among qualifying relievers. Smith was a much more worthy recipient of those pivotal innings as he posted an 11.7 RE24, making him one of the better relief pitchers in the game last season.
The usage of Smith is the crux of the recent trend toward teams acquiring elite relievers to use in a “setup” capacity. Setup is in quotes here because setup is not, like “Closer,” always clearly and restrictively defined; managers have more leeway in how they deploy these arms. This means that pitchers like Wade Davis (28.46 RE24), Dellin Betances (26.35), or Darren O’Day (17.99) could be leveraged when their teams needed them most, regardless of whether it was a specific inning or the margin of a lead was exactly 1-to-3 runs.
Enterprising front offices are willing to pay a lofty price to shorten games and acquire a Late-inning Relief Pitching Assassin (LiRPA) who can be deployed as needed when the game is on the line. More often than not, these pitchers are going to be working the eighth inning, but it’s not a hard and fast rule like it is for closers. Darren O’Day for example, was used in the sixth, seventh, eighth, ninth and 10th innings last season. By RE24 he has been one of the five best relievers over the past three seasons, and (strange as it is to say this) that hasn’t stopped Buck Showalter from using O’Day whenever he felt that O’Day was most needed.
Below is a table of each team’s best pitcher by inLI, and his team rank for RE24 (among qualified relievers[ii]; excludes pitchers who changed teams). This table helps clear up whether or not the best pitchers are pitching in the highest leverage moments:
Just about half—14 to be exact—of the teams in MLB have their best reliever pitching in the highest leverage opportunities. In some cases, such as that of the Angels and Cesar Ramos, the most effective pitcher on the staff in 2015 actually came in near the back of the pack in terms of inLI.
The majority of the pitchers in the above table are closers. This makes plenty of sense. More often than not, closers are pitching late in close games, which leads to high leverage scores. They closer isn’t necessarily the best pitcher in the bullpen, though they typically are among the best.
Still, there are exceptions. Dellin Betances led the Yankees in inLI despite spending most of the season as Andrew Miller’s setup guy. This is precisely the opportunity for teams willing to acquire a LiRPA.
The reality is that saves still pay throughout arbitration and free agency, so teams are still beholden to that bullpen structure in order to keep players from revolting. We can also debate whether or not relievers having set roles helps with their routines, which is another reason why traditional bullpen structures aren’t going to be discarded haphazardly. Knowing that, it behooves smart teams to acquire two elite pitchers who can be deployed in tandem.
One pitcher, a traditional closer, is brought in to end games and/or extend extra innings contests. The second, the LiRPA, can be used whenever the manager sees fit. Perhaps they come in during the sixth inning of a one-run game when the starter has two on and nobody out. Perhaps they simply pitch the eighth inning prior to the closer coming in. Regardless, the manager has more leeway with their deployment since non-closer roles are inherently more flexible—assuming, of course, that the gradual creep of bullpen “roles” doesn’t prejudice the “eighth-inning guy” against this type of usage.
The Yankees have Dellin Betances and Andrew Miller[iv]. The Red Sox have Craig Kimbrel and Carson Smith[v]. The Orioles have Darren O’Day and Zach Britton[vi]. That’s just the AL East, and we’ve already listed six of the top 15ish relievers in baseball.
The Dodgers attempted to replicate the concept before scuttling the Aroldis Chapman deal. The Diamondbacks could have their own dynamic duo if Josh Collmenter, who would complement Brad Ziegler, improves[vii] after completely transitioning into the bullpen next season.
There have long been cries to get rid of the save stat and re-work MLB bullpens. Those cries are likely to go unanswered, at least for the foreseeable future. Smart teams are finding a way around it by acquiring a second elite relief pitcher who their manager can have more flexibility with. A smart manager can reap the rewards of non-traditional bullpen management without enraging all of his players in the process. Acquiring LiRPAs seems to be the latest trend in MLB front offices, and it’s one that is going to make the end of games a lot harder for opposing teams that have to face them.
[i] Why RE24? It’s my preferred method of analyzing RP performance because it considers the situation when the reliever entered the game, how they left the game, and what happened while they pitched in it. It gives a clear and succinct view of whether the pitcher was better or worse than average at helping his team win
[ii] RE24 is a counting stat, so it’s important that we set a baseline for how much a pitcher was used so that it’s somewhat apples to apples.
[vii] Ziegler ranked 5th in MLB for RE24 among relievers, while Collmenter finished 21st.