One of the key reasons for the Twins 99-loss season last year was the implosion in the bullpen. After winning the division title in 2010, in an effort to save salary, the Twins let free agents Jesse Crain, Matt Guerrier, Jon Rauch and Brian Fuentes sign elsewhere. While I can't fault the Twins for letting the four relievers go, considering they signed for a combined total of $39.25M in guaranteed money, quite a bit considering the quartet combined for a middling 3.81 ERA and 4.14 FIP in 2011.
I don't bring this up to beat a dead horse and complain about last year, but rather to look forward to 2012 and what we could expect from the bullpen this season. Unfortunately, it appears many of the same mistakes made in the 2011 off season, including gambling on comebacks from injury (Joe Nathan, Pat Neshek) and counting on minor league relievers to step up when performance in AA/AAA was mostly average or worse (Alex Burnett, Jim Hoey, Phil Dumatrait), appear to be repeated in 2012. After the jump, I'll go into a bit more detail, analyzing just how much the bullpen hurt the team in 2011 and look at how 2012 projections compare. But I'll warn you in advance, the results are not pretty...
First, how important is the bullpen to winning games? While there are many views regarding how much money a team should throw at closers and elite setup guys, it's clear that good pitching out of the bullpen can make the difference between a middle-of-the-road club and a division winner. Looking at win probability added (WPA) over the course of an entire season, there appears to be between a 10 and 15 win difference between the best and worst bullpens in Major League Baseball. During the 2010 season, the Twins were near the top of baseball:
|1. San Diego Padres||+8.42||2.81||2.97|
|2. Tampa Bay Rays||+6.20||3.33||3.76|
|3. Texas Rangers||+5.92||3.38||3.99|
|4. New York Yankees||+4.72||3.47||4.06|
|5. San Francisco Giants||+4.71||2.99||3.41|
|6. Minnesota Twins||+3.76||3.49||3.91|
|28. Florida Marlins||-2.04||4.01||4.08|
|29. Seattle Mariners||-5.51||4.23||4.40|
|30. Arizona Diamondbacks||-8.37||5.74||5.09|
And, as we all painfully lived through last year the Twins were near the bottom in 2011:
|1. New York Yankees||+9.11||3.12||3.65|
|2. Boston Red Sox||+8.59||3.67||3.48|
|3. Detroit Tigers||+6.44||3.93||3.98|
|4. Atlanta Braves||+6.42||3.03||3.19|
|28. Minnesota Twins||-3.38||4.51||4.43|
|29. New York Mets||-3.60||4.33||3.86|
|30. Houston Astros||-5.21||4.49||4.38|
Comparing 2010 to 2011, we see a total drop of just over seven points (7.14) in win probability added. In other words, we can attribute 7 additional losses to last year's performance just based on the drop-off in bullpen performance. Of course, this is less than one quarter of the 31 losses between a 94-68 and 63-99 ball club, but it's a significant drop nonetheless. As we would expect, we see a similar drop of seven games in the standings in the bullpen's win-loss record between the two years:
By comparison, the Arizona Diamondbacks were a virtual mirror image of the Twins, going from 65 to 94 wins in 2011. In their case, the bullpen went from a league worst -8.37 WPA (5.74 ERA) to 10th in the league, +3.24 (3.71 ERA), an 11.61 increase in WPA, or over a third of the jump from 2010 to 2011.
So what can we expect from the Twins bullpen in 2012? At this point, I consider Glen Perkins, Matt Capps, Brian Duensing, Alex Burnett and Anthony Swarzak to be virtual locks for the bullpen, with Lester Oliveros, Kyle Waldrop, Carlos Gutierrez, Terry Doyle and others in the mix. Looking at the various projections available on Fangraphs, it does not paint a pretty picture:
|Player||ERA (ZiPS)||FIP (ZiPS)|
(*) ZiPS projects Brian Duensing to start 20 games in 2012. Injuries may force this, but if he remains in the bullpen full time, used mostly against left handed hitters, expect him to improve his ERA by quite a bit. However, if he ends up pitching largely in LOOGY situations, we'd also be talking about fewer innings pitched, so a double edged sword.
As the table above shows, projections are not kind to the projected Twins bullpen in 2012. I used ZiPS, but one could just as easily use one of the other systems such as PECOTA, Marcel, etc and see similar results. Given these projections, it appears to me that the Twins would end up with a bullpen ERA much closer to the mid-4s, as we saw in 2011, than in the mid-3s from 2010. Of course, I would not expect a pitcher like Gutierrez to stick around for long with an ERA in the high-5s, and there's a decent chance that one or two of the rookies at least give average if not better performance. But at a glance it looks like there will be quite a few replacement level innings out of the bullpen this season.
Bottom line, we have to hope that the starters eat quite a few innings this year, resulting in quite a few games where Duensing, Perkins and Capps can take care of the 7-8-9 innings. But I'll get into the starters next week.
What do you think about the bullpen this year?