Billy Bean’s recent extension of reliever Sean Doolittle‘s contract has left many sabermetricians scratching their heads. The prevailing wisdom in sabermetrics is that relievers are easily replaced and therefore not worth the millions they are often paid. Because Billy Bean has been crowned the sabermetric champion, many baseball fans are confused by his willingness to divert resources to a seemingly expendable position.
The belief that good relievers are easily replaceable is actually a failing of mathematical models that rely on an extensive amount of data to provide useful conclusions. Like with any other position, players good enough to excel at the major league level are in short supply.
Most relievers are volatile and inconsistent. They have good years and bad. Their play time is determined by their success. They bounce up and down between the majors and minors, get cut, change teams.
On rare occasions, a reliever is steady and good, giving one solid outing after another, year after year. On rarer occasions, a reliever steadily dominates opponents year after year. Greg Holland is such a pitcher. He passes all the tests of what makes an elite reliever.
The most important thing about relievers is that they keep a lead. It’s the whole reason the rather unhelpful “hold” and “save” statistics were created. Of course it’s always good to prevent runs, even when losing, but, primarily, these guys shouldn’t blow games.
To prevent blowing as many games as possible, a bullpen must remain consistent–be good every single day. Teams don’t want some guy coming in and blowing six or seven leads a year and posting a 4.00+ ERA in a down year. Much like the Ace in a starting rotation, the best relievers don’t have down years.
Because they’re relievers and don’t have to worry about facing a lineup multiple times per outing or throwing 100+ pitches per outing, good relievers should have a much better ERA than starters, who have to pitch 170 innings or more in a season. Very good relievers generally have ERAs under 3.00. They also should pitch a minimum of 50 innings per year, otherwise, they might become a needlessly expensive roster spot.
When teams look for “elite” relievers, they want to see a track record of success. Rookie pitchers who are called up from the minors, who some analysts believe can easily replace major league relievers, are unpredictable. Some strategists believe that teams should give a bunch of relievers a try until they find one that sticks. However, this could take several tries and the young relievers could blow several games while the club tries to determine if he has the stuff to make it in the big leagues. The Toronto Blue Jays proved that a team can form a bullpen in this manner in 2013, but not before they cycled through 21 relievers.
Two things about this strategy have negatives.
The first should be clear. Cycling through that many pitchers takes time away from more productive pursuits. Time is money, and taking time to manage a roster costs money.
For better or worse, it’s a chief tenant of the Royals organization led by General Manager Dayton Moore. In negotiations especially, Moore sets out to get a player and does not waste time haggling. If negotiations do not progress expediently, he moves on to the next guy. Wasting time on unproductive activities can be costly, as the Rangers proved after the 2012 season while negotiating with Josh Hamilton and Zack Greinke, neither of whom they signed.
The Rangers haggled long and hard before losing both players to other teams and watched many opportunities pass them by in the mean time. They entered the season with an oft-injured Lance Berkman at DH, a struggling David Murphy in left field and a marginally efficient platoon in center-field. Their starting pitching depth was so shallow that rookie Nick Tepesch notched 93 innings. They missed the playoffs by one game.
Taking one fourth of the season to determine the best bullpen options will cost the team some wins and can distract a front office from improving the team in more concrete ways. (That being said, testing a slew of relief pitchers in a year that has already tanked, or was predestined to tank, is a great idea.)
Second, teams are looking for a relief pitcher they can count on, someone they don’t have to worry about, a pitcher who has been good for the past three seasons or so. GMs want as much of a sure thing as they can get. They don’t want to enter a season with a potential hole in their roster when they can, instead, count on their guys to replicate their previous success.
General Managers should be able to count on their top relievers to consistently deliver at least 50 innings per year. Their ERAs should be less than 3.00, and not only that, they should never balloon over 3.00 for a season. A pitcher can’t do much more than that to set an organization’s mind at ease.
The rarity of a reliever meeting those qualifications puts them in the upper echelon of bullpen categories: super-relievers; the elite; bullpen aces. How rare are they, you ask?
From 2011 through 2013, out of the 834 pitchers to make relief appearances in the major leagues (about 500 were regular relievers for a time), the list of relievers that have accomplished this feat of consistency and dominance is just twelve names long:
Tables provided by Baseball-Reference.com
Koji Uehara and Sergio Romo just missed this list with too few innings. We could easily add them without diluting the list’s integrity.
Of all active relievers, only 40 relievers have ever pitched 50 innings with a sub-3.00 ERA three times, much less three times in a row. The Royals have one of the 12 guys who have in Greg Holland.
As a fun exercise, I also looked at the next tier talent of relievers. The guys who aren’t elite, but are consistently solid. Only thirty pitchers have managed to get into the next tier, where the pitcher has thrown at least at least 45 innings and had an ERA under 4.00 in each of the years from 2011-2013. Most pitchers who have an ERA north of 4.00 don’t get a lot of playing time, so having a reliever with an ERA under 4.00 removes a lot of uncertainty about replacement relievers during the season. In addition to the Elite relievers above, Royals relievers Tim Collins and Aaron Crow also make the cut; Louis Coleman just missed. About 100-115 pitchers a year accomplish this feat, but rarely do they keep it up every season.