Steve Cishek: Why Not?

Steve Cishek was a top-ten reliever this year. Steve Cishek! He’s supposed to be selling seashells by the sea shore, and here he is providing valuable stats for a low cost. Provided what we know about relievers on bad teams, why were fantasy owners nervous about Cishek going into the season? And did the last 70 innings solve those problems? If so, why not take Steve Cishek as a top-tenner this year?

After a 2012 season that saw Cishek take over the closer role in Florida, perhaps we should have been more excited. He had struck out more than a batter per inning (24.7%), given up more than half his contact on the ground (52.4% GB), and his home park had helped him suppress homers (5.8% HR/FB) to the point that despite an iffy WHIP (1.30) built on bad control (4.1 BB/9, 10.6% BB), he had a sub-three ERA on the season. The team didn’t really have great options other than perhaps unrefined youngster A.J. Ramos, and he seemed primed to return good value for a cheap cost.

Well, he did. He even overshot that mark by improving his control. His walk rate was actually better than league average last year (2.84 BB/9, 7.8% BB). Looking back at his minor league career, this was a thing that happened — he alternated years of worse-than-average control with years with good control then, too.

That doesn’t make much actual sense though — why would control come and go like that — unless we can figure out the why behind it. Looking at his per pitch numbers on Brooks Baseball, there are some tempting numbers to point to. He gradually improved the ball rate on his four-seam from 42.4% to 36.5% from 2011 to 2013! That’s it! Except that he throws the four-seam, on average, less than nine percent of the time. It’s that much-less sexy improvement on his sinker — 34.3% balls to 31.6% balls — that’s more likely the pitch improvement that led to fewer walks. He’s thrown 1664 sinkers against 265 four-seamers anyway.

If better fastball control was all it took, and he’s improved in that category for two straight years, all while retaining his 92+ mph velocity, it’s all systems go going forward, no? Maybe.

And now, despite emphasizing the pitches he throws the most for the first point, we’ll have to look at the pitch he throws the least for the next point. After all, he never throws the change to righties, and throws it almost 10% of the time against lefties. That’s probably because his primary pitches, the sinker and the slider, generally have bad platoon splits. (And specifically too — Cishek’s slider’s whiff and grounder rate drops below average against lefties, while his sinker retains a decent whiff rate while losing grounders (45%).) The changeup is bad, though. It doesn’t get average whiffs (12%) or grounders (36%). It has a lot of horizontal movement, but only about an inch more drop (and seven mph slower) than his sinker. It’s too similar to his sinker, most likely.

Look in Cishek’s splits, and you’ll see some evidence of splits. His walk rate against lefty batters almost doubles (11.3% from 6.7% against righties, career), and his ground-ball rate tanks (61% to 44%). So far, his low home run rate has helped mitigate the problem. And his home park is the second-hardest place in the big leagues for a lefty to hit a home run, so that might continue while he calls that monstrosity home.

Using a closer’s arsenal to predict future closing changes hasn’t worked well for me in the past, and this might be a clue why. It’s not a big deal until it’s a big deal, in other words. Cishek may not have a great way to get lefties out, but it works in Miami. If I were buying him this season, I’d be more worried about new arrival Carter Capps — whose higher velocity and strikeout rates are things that are usually associated with closer change (FG+) — and the fact that any desirable asset in Florida is a risk to get moved at any time.