Of shutdown, meltdowns and blown saves: the 2013 Diamondbacks bullpen

It didn't take long for the Diamondbacks bullpen to become an issue this season. The first blown save of the yeat happened in the third game, when Tony Sipp couldn't hold a sixth-inning lead against the Cardinals. By the time the year was over, there would be 28 more, giving Arizona a tally surpassed only once since 2004 in the National League [the 2008 St. Louis Cardinals had 31] And yet, 19 of those were in games the Diamondbacks still won, albeit taking a bit longer to do so - quite a lot longer, in some cases. That's the most victorious blown saves since anyone started counting the save statistic.

At times, the problem seemed to be as contagious as bad tacos, and just as distressing to my intestinal tract. In one 15-game spell, from April 17-May 1, the Diamondbacks bullpen blew leads in eight separate contests. And the 2007 slogan "Anybody, anytime" took on a new, ironic meaning, because over the whole year, no pitcher ever blew two saves in a row. Oh, sure: there were names which triggered a Pavlovian wince when they were announced: The Heath Bell Experience led the way with seven, but David Hernandez blew six and J.J. Putz had five. But ten different relievers were charged with blown saves over the course of the campaign.

If you want a demonstration of reliever volatility, I give you the 2013 Diamondbacks. After his disastrous first outing, Bell had a 2.22 ERA over his next 25 games. But he then allowed home-runs in five consecutive relief appearances, a streak not surpassed in the NL while Arizona has had a team. Then, over the next 25 games, it was back to a 1.82 ERA. Another period of uber-suck followed - six games, 4.2 innings of work, nine earned runs - before he finished the season with a scoreless streak. Put another way: Bell had an ERA of 1.75 this year, if you cross off Opening Day, two weeks in June and maybe ten days in August. But over those three spells, Bell's ERA was 19.00.

That sheer inconsistency was perhaps the core of the problem. You could choose just about any of our late-inning, high-leverage guys, save Brad Ziegler (who consistently Ziegled, and didn't have an ERA above 3.60 for any calendar month), and do the same. There were sections of sheer dominance, where it seemed Hernandez or Putz  would never allow another earned run. And yet, like flicking a switch, there would also be stretches where it seemed they would never retire another batter. Now, I know relief volatility is very, very real: but the roller-coaster this year seemed to be stuck in Final Destination 3 mode for Arizona this year. The louder you scream, the faster we go...

We've had our share of heart-attack bullpens in the past: the one which comes to mind immediately is the 2007 version led by Papa Grande. They certainly provided us with their own meltdown moments, e.g. coming within a bloop single of blowing an eight-run lead in three innings, and had a one-run record of 32-20, very similar to this year's 34-21. But overall, they were probably "better": relative to league bullpen performance, they had an ERA below average, and certainly, they had barely half as many blown saves (15) as the 2013 model. This year, it was like Forrest Gump's chocolates when the bullpen gate opened. You never knew what you were gonna get.

Another lesson to be learned from the season, is perhaps also linked to volatility - the danger of overpaying for relief performance, which seems a lot less reliable than starting pitching or position player output. The chart above shows our relievers with 20+ IP: if you can find any real correlation between salary and value, you're better then me. The players receiving at or close to league minimum were all in the middle, with the higher-cost ones exclusively occupying the extremes. Bell was worth a whole win less than waiver-wire pick-up Harris: somehow, the fact the Marlins paid $4m, for him to perform below replacement level for us, isn't much comfort.

Everyone wants someone like Mariano Rivera, who hasn't had a season ERA+ worse than 144 since going to the bullpen in 1995. However, he's very much the exception rather than the rule. There are only two National League closers with more than 15 saves in each of the last three seasons (Craig Kimbrel and Huston Street). Even expanding our selection beyond closers, only 16 relievers in the entire majors have thrown even 50 innings with an ERA+ of 115 or better, in each of the past three years. I'm inclined to think the $12 million we'll pay Putz and Bell next season, could be used more productively elsewhere.

One area that was particularly disappointing was our left-handed options. I was hopeful, before the season began, that we would finally have decent left-handed relief, something which has been ephemeral on the D-backs, for as long as I can remember. Doug Slaten, Clay Zavada, Joe Paterson, Craig Breslow: all flew across the mound from left to right, for one good season, before fading into the baseball night, for one reason or another. With Reynolds and Sipp this year, I was hopeful we were set, and Reynolds started brilliantly. Then his season - and the next one - was ended by injury, and Sipp allowed lefties to hit at an .859 OPS.

As a replacement for Reynolds, we traded Ian Kennedy for Joe Thatcher, but the early returns on that weren't much better. While small sample size obviously applies, and he wasn't helped by a bloated BABIP, southpaws clobbered Thatcher to the tune of .368/.500/.579. It leaves me wondering how much of his success with the Padres was perhaps the result of Petco's friendly confines, and as a second-year arbitration eligible player, he'll likely be earning a significant amount next season too. Hopefully, he can live up to his previous performance, rather than the shaky start we saw post-trade.

The other key question for next year is the closer's spot. We saw three main incumbents this year: first Putz, then Bell and finally Ziegler. The last was the only one to perform well in save situations, with a 2.19 ERA compared to 4.20 for Putz and 4.61 for Bell, but even Brad himself said, "Everybody knows, in a perfect world, that’s not my role on this team." It'd certainly be preferable to have him return to a "reliever without portfolio" position, coming in whenever we need a big ground-ball: in some ways, it's even more important than the largely artificial, "ninth inning with a lead of three runs or less" job.

I suspect we'll see Putz back as closer on Opening Day next season, and Hernandez seemed to have recovered his mojo after the spell in Reno; if that continues, he'll be a great set-up man, with Bell in the seventh and Ziegler used as necessary. Collmenter will be long relief and Thatcher as left-handed specialist. That will only leave one spot open, assuming the standard seven-man bullpen and no injuries: Will Harris and Tony Sipp appear the likely candidates, and Harris certainly performed better this year: but it's possible Gibson prefer two left-handed options in his 'pen. We'll see next spring....