If you've read and retained much from the sabermetrics movement, then the variability of bullpens is quite possibly something you're familiar with. As the standard reliever analysis reads, these guys are usually in the bullpen because they aren't good enough to be starters and because the true talent bar isn't set all that high, they're prone volatile results. Of course, the discussion could be more nuanced depending on the pitcher at hand, but that's essentially the gist of it.
It can be difficult to get excited about relievers or particular bullpens when you understand these principles. Sure, you catch a great outing or two, but you start to expect that something's about to go awry and the regression to the mean is looming like a dark cloud waiting to pour on your team in an unfortunate way. A current candidate for applying this line of thinking might just be the Milwaukee Brewers.
The Brewer's upgraded bullpen has been stellar through the team's first 12 games. Francisco Rodriguez has been perfect thus far as the team's closer, showcasing a nasty breaking ball, but several other solid contributors back him up. Jim Henderson, Tyler Thornburg, Brandon Kintzler (who just hit the 15-day DL) and Will Smith are proving themselves as capable relievers. Thornburg has been excellent as a reliever, Kintzler has been a ground ball machine, Smith is excellent against lefties and the results show this collective dominance: Milwaukee has the second-highest bullpen WAR on the young season and the team's relievers account for 50% of their cumulative pitcher WAR. Needless to say, the results in 2014 have been outstanding thus far, as you can see below:
These five guys have been nearly perfect over the first two weeks of the year. There are several reasons for Milwaukee's 10-2 start (best in baseball through Sunday), such as a healthy and effective foursome of Ryan Braun, Carlos Gomez, Jonathon Lucroy and Aramis Ramirez and five reasonably good starts between Yovani Gallardo and Matt Garza, but there's no doubt that the bullpen domination has been a spectacular aid to the Brewers' early winning.
Some serious regression should be in order, however, as we discussed at the outset. There's clearly no way the bullpen can continue on its current pace as such production would be otherworldly. Because recent performance is the best indicator of future performance, let's examine last year's results from the same relievers. As a bonus, we'll include the ZiPS rest of season projections, too.
|2013 & 2014 ZiPS (R)||K/9||BB/9||HR/9||BABIP||FIP||WAR|
|2014 ZiPS (R)||10.38||3.42||1.01||0.313||3.52||0.3|
|2014 ZiPS (R)||10.84||4.15||0.97||0.309||3.73||0.3|
|2014 ZiPS (R)||7.74||4.10||1.30||0.306||4.89||-0.4|
|2014 ZiPS (R)||6.61||2.68||0.74||0.307||3.78||0.1|
|2014 ZiPS (R)||8.09||3.10||1.15||0.311||4.21||0.1|
When we look at what's happened in the past and what is forecast for the rest of the year, some things start to really stand out. For one, each pitcher, aside from Henderson, is currently benefitting from an unsustainably low BABIP. Small sample sizes are known for distorting these kinds of numbers and while the Brewers can play some decent defense, this is going to change and probably sooner rather than later. The candidate most likely for severe regression, according to ZiPS, is Tyler Thornburg, who thus far, has been electric. Keep in mind that ZiPS has the least amount of knowledge of Thornburg given his short MLB track record and that some of it's projections are as a starting pitcher, not as a reliever, which does change the equation somewhat.
The other major note is that none of these five Brewers relievers have surrendered a home run yet in 2014. To be fair, ZiPS doesn't necessarily suggest that these are abundantly homer-prone pitchers, but they've combined for 28.1 homerless innings so far and like their BABIP results, this cannot be sustained. With their projections, we'd expect that they'd have given up three or four longballs thus far, but alas, none have been taken yard yet. Considering that Rodriguez (37.6%), Smith (46.1), Henderson (26.8%) and Thornburg (38.2%) all posted low groundball rates last year, this trend toward limited home runs may be about to be flipped on its head.
What I don't want this to sound like is a criticism of what is a pretty good bullpen. Make no mistake, Milwaukee's relief corps can be better than most. But they're not this good because no one's this good. It just doesn't exist over the course of a full season. We're all looking for the next breakthrough, but keeping small samples in mind, this Brewers bullpen is about to fall back to earth. It's a good thing they've got the best record in baseball, because when volatility strikes the bullpen, they're going to need to rely on the cushion they've built to remain in the race for as long as they can.
All statistics courtesy of FanGraphs
. . .