The Crain Drain
More on the diminishing K/9 and what it means

Aaron Gleeman wrote an article for The Hardball Times today on a subject I’ve commented on a couple of times: Jesse Crain’s vanishing strikeout rate. While I agree with his main premise, I think he undervalues the importance of Crain’s lost strikeouts.

Crain is a unique case in that he’s near the top of almost every leaderboard for relief pitchers, sometimes for the best and sometimes not. For one, he has a narrow lead over Joe Nathan for 12th in the majors in Relieve Expected Wins Added at 2.397, also good for 7th in the AL. He has actually been even better than that, but many of those in front of him face higher-leverage situations as closers, or even primary set-up men, and WXRL takes into account the importance of the situation based on game state data. On top of that, his first order pitching stats look shiny, too. For instance, he’s sporting a 2.49 ERA and a 9-1 record, enough to make Dick and Bert salivate over his new two-seamer.

But here’s what I wrote about him in my first half review column, explaining why he wasn’t a choice for one of the “Happiest Surprises:”
Jesse Crain: The wins and the “vulture” tag are fun, but don’t be duped into thinking that K rate is meaningless. In the show, he’s struck out 29 to 23 free passes in his first 68 2/3 IP. Sure, it’s possible that he has undermined mountains of data that suggest the pitcher controls Ks, BBs and HRs but has little control thereafter, in the same way that it’s possible that I could have a successful major league career. Someone needs to talk that Crash Davis out of him. Strikeouts aren’t boring, and even if groundballs are more democratic, I’ll take a fascist Pedro Martinez any day.

And I’ll stand by that comment today. Crain cannot keep up his current level of success over the long term without improving his K/BB rate, mainly by bringing up the strikeouts. Someone compared him to Mariano Rivera in surrendering less hits than expected regardless of defense. Fundamentally, though, surrendering less hits is a function of striking batters out. Remember Voros McCracken’s work on BABIP, that there are three true outcomes that are defense independent: walks, strikeouts and homeruns. Crain has done exceptionally well limiting HRs, certainly a result related to his shift from a FB to a GB pitcher. His BB rate has been solid, not on par with the starting rotation, but about ¾ of a BB better per 9 than his contemporaries, Nathan and Rincon. But he stands out due to his ridiculously low 3.26 strikeouts per 9IP.

Some argue that he has improved at making his pitches harder to hit, jamming batters and preventing them from making decent contact. Indeed, if inducing infield flies is a repeatable skill, then it could have an impact on a pitcher’s BABIP, but in Crain’s case, a) he’s not inducing more pop-ups (17% of all flyballs, in line with league average), but more GBs, which don’t meaningfully prevent hits, and b) presumably, pop-ups are a similar result to strikeouts, where the hitter badly misses the pitch, but makes little contact instead of none. They should increase together.

Also, pitchers can control G/F ratio, but not where the ball goes on the field. Not even Pedro can pitch to his defense. GBs are more desirable than FBs as they don't turn into XBHs as often, but both become hits due to factors outside of the pitcher's control. Obviously, even if G/F ratios stay the same, BABIP fluctuates wildly, which undermines the basic principle that G/F ratio has much to do with balls turning into hits. Big problem: Crain has the 6th lowest BABIP among pitchers over 40 IP at .211 (ironically right next to Rivera). There are lots of good pitchers near the top of the list (Prior, Wagner, Harden, Clemens), but also lots of undistinguished ones (Matt Wise, Dave Riske, Cliff Politte). And if you think his improved G/F ratio is the causal factor that reduces his BABIP, consider that his 1.10 G/F ratio pales in comparison Silva’s sterling 1.68 ratio, but his BABIP is .089 lower, .211 for Crain to .300 for Silva. Just throwing sinkers doesn’t prevent hits. If anything, it prevents them from turning into HRs, but Crain is experiencing a beneficial phenomenon beyond that explanation. All in all, once the ball hits the bat, the pitcher has pretty much nothing to do with where it goes next, as long as it stays in the park, so Crain’s low BABIP is more luck than skill, and the its persistent low value over two seasons is a statistically anomaly.

I agree with Aaron’s argument that he’s stretching his defense awfully thin, but I think it goes beyond just walking a tightrope. As soon as Crain’s BABIP catches up with him and starts regressing towards league average, he’s due for a pretty harsh wake-up call that will result in a swelling ERA.