Jeff Locke rides the DIPS rollercoaster

People ask me* what I do in the winter when there's no baseball. I'll tell you what I do. I analyze and write about something baseball-related, or I watch football. This weekend, after trying and failing to accomplish the former — finding a topic is hard — I gave the latter a whirl. That led me to witness one of the most egregious slip-ups of the NFL season: Vikings kicker Blair Walsh shanking an easy field goal, which allowed the Seahawks to win 10-9. As a fan of the Ravens, I can empathize with the current suffering of the Minnesota faithful (sorry, Drew Magary); still, I had to laugh at their team's plight.

*No one has ever asked me this.

Of course, Walsh didn't err alone. Many people noted that his holder, a man by the name of Jeff Locke, didn't set the ball up properly after receiving the snap. This blunder — which Locke owned up to — reminded me of another man of the same moniker, who's also earned some criticism recently. From there, I looked into several facets of the latter player's case, leading me to an interesting discovery.

Jeff Locke, the Pirates pitcher, came onto the scene in 2013 with a solid 98 ERA- in 166.1 innings. That belied more mediocre peripherals, though: he had posted only a 110 FIP-. In the next year, the gap between the two shrunk a bit, as he notched a 111 ERA- and a 121 FIP-. Then came 2015, when the entire thing flipped around — his ERA- exploded to 120 while his FIP- progressed to 104. All in all, Locke has had a crazy few years:

*Out of pitchers with at least 100 innings.

Why did he suddenly switch from an underperformer to an overperformer? His BABIP, which jumped from .278 in both 2013 and 2014 to .312 in 2015, seems like the culprit. Perhaps he allowed hitters to hit balls against him, causing more of them to fall in for hits. Yep, this is an open-and-shut case — nothing else has affected him!

We'll begin with Locke's contact ability, to which the opposite has happened. From 2013 to 2014, he gave up a soft-hit ball 16.9 percent of the time and a hard-hit ball 30.1 percent of the time. Those marks improved to 21.3 percent and 27.0 percent, respectively, last season. Even after adjustments for the league environment, Locke still comes out ahead: His hard-hit rate went from one percent worse than average to six percent better than average, and his soft-hit rate leapt from two to 15 percent better.

Nor did Locke's batted-ball distribution shift much. He saw his ground ball rate fall one percentage point, from 52.0 percent to 51.0 percent; since balls on the ground go for hits more often than balls in the air do, this should theoretically help his BABIP. Although his line drive rate shot up to 24.1 percent from 20.5 percent, I wouldn't read too deeply into that because line drives (a) correlate very poorly year-to-year and (b) don't correlate that well to BABIP. Batters couldn't square Locke up as often, and they made the same kind of contact when they did put the ball in play, meaning he likely doesn't bear the blame for his increased BABIP.

So what is the main cause for the increase in BABIP? Pittsburgh's defense declined in recent years. According to DRS, it saved 68 runs in 2013, then 36 runs in 2014, then seven runs in 2015. UZR sees the decline a bit differently — in its eyes, the main dropoff came between 2013 (4.4 runs) and 2014 (-40.3 runs), followed by some progress in 2015 (-20.8 runs) — but the concept remains. The slick fielding that defined earlier Pirates clubs has disappeared, and in one place particularly:

As a heavy ground-baller, Locke relies on his infielders to convert balls into outs. When they fail to complete their end of the bargain, he suffers, as we've seen. Pittsburgh's infield shift strategy somewhat confounds the defensive numbers, but they have been employing this strategy for several years now.

This likely doesn't explain all of Locke's poor play last season. Some of it comes from the inherently fluky nature of balls in play. While many have justifiably critiqued it since its initial publication, Voros McCracken's initial work on defense-independent pitching still rings true 15 years later. If an extra few balls find the glove of a defender, that pitcher's BABIP will fall; if those balls instead dart away from fielders, then his BABIP will rise once more. The men on the mound can control only so many elements of this game, and ERA often doesn't give them the credit (positive or negative) that they deserve.

What does all of this mean? For Locke's 2016, we should expect normal results. ZiPS foresees a 4.15 ERA and 4.11 FIP, while Steamer predicts a 4.01 ERA and 3.92 FIP — both reasonable projections. Of course, he could under- or overperform significantly again, but we should assume he'll regress to an average level, as projection systems are regressed, mean-based systems. Neither the low-BABIP Locke of old nor the high-BABIP current Locke had the peripherals to back up those unusual marks, so they likely won't happen again.

This year, Locke may not pitch much for the Pirates, who sorely need a better rotation and could move him to the bullpen. Whatever his role, he should take solace in what he's achieved to this point. The high 2015 ERA doesn't reflect his play, and although FIP has its flaws, it can better testify to his improved performance. We'll just have to see what this crazy ride will bring for Locke in the coming season.

. . .

Ryan Romano is an editor for Beyond the Box Score. He also writes about the Orioles on Camden Depot (and on Camden Chat that one time) and about the Brewers on BP Milwaukee.