No reason, no complaints replacing Nunez with Hensley

Consider the following two relief pitchers (all numbers from 2010 with similar innings pitched and batters faced):

Not much difference between the two, right? One seems to have been a bit more lucky on his fly balls than the other, but both have been inducing their fair share of grounders. Both have struck out a lot of hitters while walking a below-average number of batters along the way. Their resulting xFIPs are similar, and Baseball Prospectus’ SIERA (an ERA skill estimator similar to xFIP) produces numbers even more similar (Pitcher #1 at 3.00, Pitcher #2 at 3.15).

Who are these guys? From the title, it should be obvious that one is Leo Nunez and the other is Clay Hensley, the only two Marlins relievers who have done anything worthwhile. However, what should not be so obvious is which one is which. It turns out Pitcher #1, who has been slightly better skill-wise (normalizing for random variance) but worse performance-wise, is Nunez, while #2 is Hensley.

The fact that you cannot easily differentiate between them is particularly interesting given the fact that the Marlins just recently demoted Nunez in favor of Hensley for the closer role. Why has Edwin Rodriguez made such a move? The answer lies in the combination of high leverage innings and small sample sizes.

The issue that Rodriguez is seeing in Nunez is probably just a combination of good old fashioned regression to the mean and a little bit of bad luck. Since his first outing in August, Nunez has allowed 12 runs in 11 2/3 IP, leading to a 7.70 ERA. I discussed his “bad play” in August and attributed it to bad luck and regression before as well; at the time his strikeout and walk rates between his first half of the year and his second half “struggle” (encompassing July and August) had remained mostly the same. If the guy is striking out and walking the same number of guys, I’m going assume his stuff is msotly the same.

Lately, however, he has not been doing the same thing. He has had a stretch from August 19 to September 3 during which he has walked five guys out of 30 hitters faced, struck out only six of those batters, and allowed a whopping four home runs. This is only in 30 batters and 5 2/3 IP! Despite an otherwise fortunate .267 BABIP, Nunez has been battered for six runs.

Rodriguez is seeing this and saying “something ain’t right.” I’m seeing this and saying “pitchers can have five bad innings and still be the same.” If I recall correctly, no one skipped a Josh Johnson start because he had a lousy previous one. These sorts of things just happen. The fact that it occurred during a half-month stretch is irrelevant; Nunez only faced 30 batters, or the equivalent of one start for a starter. It’s too small a sample size to be using to determine a player’s role on the team.

Luckily, the Marlins are saved for two reasons. Yes, the team is demoting Nunez despite the fact that this is his best season of his career, but unlike last season, the team actually has options this year. Last season, Nunez was significantly worse (xFIP of 4.41; SIERA of 4.02) but was allowed to stick around because the next option was Kiko Calero, a pitcher the Marlins apparently didn’t trust. This year, it seems like the Fish are comfortable putting their trust in Hensley despite his similarly journeyman history. Before this season, Hensley had only one successful season in the majors, and that was as a soft-tossing starter who kept the ball on the ground. This season, the ball is still mostly on the ground for Hensley, but the transition to the pen and the change to more changeups has upped his strikeouts significantly.

The other reason for the relative unimportance of this move is that Rodriguez and company seem to recognize that this is likely a short-term move. According to the linked MLB.com blurb, both Hensley and Rodriguez seem to accept the move, short-term or otherwise:

"“He’s looking better. At some point, we’re going to close with him,” Rodriguez said. “Hensley is a very good option. Having those two guys able to close guys, it will be good.”Hensley has appeared in 55 games and he’s struck out 66 in 60 2/3 innings.“I think Nunie is still the closer, but I’ll take advantage of the situation,” Hensley said. “There may be a little more added pressure, but I think there is as much pressure in the eighth. You don’t want to blow it. I feel like I’ve been facing a lot of the meat of the order all season. Instead of going to the dugout, I get to shake the catcher’s hand.”"

There was no reaction from Nunez, but he can’t be too disappointed with those words. If the Marlins are smart, they will see Nunez as the future in the ninth inning for the Fish, at least for the next few seasons. The current projection systems are not terribly reliable in terms of projecting Hensley going forward, as both CHONE and ZiPS are still working with him as mostly a starter, but according to their numbers, Nunez does project about half a run or more better going forward. Consider the arsenals involved here as well. Nunez is more of a classic hard-thrower, with a fastball averaging in the mid-90’s (94.1 mph according to BIS data) and a high-80’s changeup as his only weapons. Hensley, on the other hand, is still working with a high-80’s fastball along with a curve and changeup. While Hensley is more likely to keep his grounders because of his fastball’s two-seam effects, Nunez is more likely to maintain the high strikeout totals just based on his fastball alone.

Either way, the Marlins cannot be too disappointed with either reliever as their ninth inning guy. Though the demotion is probably unnecessary, it is also unlikely to be damaging short- or long-term.