Steve Cishek and the Framing Effect
Righty reliever Darren O’Day recently inked a four-year deal to remain with Baltimore which will earn him $31 million over four years. O’Day throws side-arm and often struggles to push his fastball past 86, but he’s undeniably one of the toughest relievers to hit in MLB right now. O’Day combines high strikeout rates with moderate walks, but as you’d expect from a soft-tossing righty, his success depends in part on his ability to generate weak contact. O’Day isn’t a ground-ball pitcher, but his funky arm angle and ability to back-door his slider to left-handed bats not only allows him to produce fly balls without too many HRs, it allows him to avoid the platoon splits that you’d figure given his pitches and how he throws them. Even a few years ago, giving a 4-year deal to anyone but an absolute no-doubt, lock-down closer would’ve been frowned upon (“they’re too volatile!” “You can just grow your own!”), and after Rafael Soriano and, to a lesser extent, Jonathan Papelbon’s deals went south, you’d be forgiven if you thought we wouldn’t see more such deals for anyone below the Chapman/Kimbrel/Jansen class, or lower AAV deals buying out free agency years for the likes of Ken Giles and Dellin Betances.
Obviously, the deal O’Day just signed – and the supportive reaction it’s received – is in part a sign of baseball’s financial health and the recognition not only that a shut-down bullpen can be critical to team success, but that some relievers really do seem to have the ability to “beat” their fielding-independent stats. That is, O’Day’s appearances come in highly leveraged situations, and he seems to have the ability to strand runners both due to his K rate and the whole infield-pops and opposite-field-texas-leaguers thing. Some may still quibble with this; giving $31m to a 33 year old seems risky, and if all relievers are volatile, the ones in their mid-30s who throw 85-88 might seem to be especially prone to it. Still, there’s no question that the market is more and more comfortable giving longer contracts to relievers. Relievers are getting more guaranteed money, over longer contracts – call it the Royals effect, after Kansas City’s bullpen-fueled run to consecutive AL Pennants and a WS Title, or chalk it up to the rising tide of MLBAM revenues or whatever, but it’s real, and it’s going to continue.
Carson Smith, as I’m sure you’ll recall, was the centerpiece in a trade for a young, cost-controlled *starting* pitcher this offseason, but you can *still* make the case he was undervalued. Smith’s 2.1 fWAR ranked 5th in baseball in 2015, the product of a great K rate and another kind of contact management: a freakish GB rate. His platoon splits, like O’Day’s, should be a problem despite the fact he throws 93, not 86: he’s a sinker/slider guy, two pitches that are generally the worst offenders in platoon split problems. Smith doesn’t backdoor his slider to lefties, he throws it in the same spot he throws it to righties, but the break and his deceptive delivery mean lefties just roll it over. Sure, sure, righties roll it over too, but lefties do so even more, and the *idea* of it – of defending against a weird slider you’re having trouble picking up out of Smith’s hand – makes his sinker more effective. Smith does it in a very different way to O’Day, but Smith too is able to pitch effectively against lefties: no lefty has homered off of Smith yet.
The Astros recently completed a trade with Philadelphia that sent young fireballing reliever Ken Giles to Houston. The Astros gave up young, hard throwing SP Vincent Velasquez (who K’d 25% of the batters he faced in the majors, and *35%* in AA last year), recently-drafted control artist Thomas Eshelman, and fly-balling back-of-the-rotation workhorse Brett Oberholtzer. Late last week, we learned that Philadelphia would *also* receive former #1 overall draft pick Mark Appel from Houston. The Phillies got quite a haul for Ken Giles, is what I’m saying. I think Wade Miley is a perfectly good #3-#4, and given his contract, that’s not a bad return, but the Giles package is categorically different. I think it’s possible that Giles may have drawn more interest from teams given his velocity, but again: Smith’s K rate was *better* despite pitching in the AL. You can hate fWAR for relievers all you want – Smith’s *FDP-WAR* was great too. Better than noted contact-manager Joaquin Benoit’s the last two years, better than Giles’, and on par with Andrew Miller’s – the guy who signed a 4-year, $36m deal last year. Smith is clearly in that Miller/O’Day class, but he makes the league minimum and, whether he wants to be or not, under the control of the Red Sox for 4+ years.
Ok, that’s a long-winded introduction to a post that’s supposed to be about Steve Cishek. The point here should be clear, but I’ll spell it out: how you feel about the M’s signing ex-Marlins closer Steve Cishek to a 2-year, $10m deal probably has a lot to do with your frame of reference for the move. On the one hand, Cishek looks an awfully lot like Darren O’Day, from the funky side-winding delivery to the domination of left-handed bats to the elevated strikeout rate. And hey, due to what may have been due to BABIP-luck and a weird but ultimately meaningless tired arm early in the year, he’ll have to take a *fraction* of what O’Day’s getting from the Orioles. The M’s bought low, and get a high-quality reliever’s bounce-back years for very little. From the other point of view, O’Day deal is completely irrelevant. The M’s needed a reliever to replace Carson Smith, and they got one, only older and with some serious red flags (Cishek’s velocity dropped by between 1-2 MPH last year), and signed him to a contract that’ll pay him roughly *10 times* what Smith would earn.
Advocates for seeing this as a great buy-low move might point out that given reliever volatility, shopping for relievers coming off of down years should be a clear arbitrage opportunity, and a way to add impact talent to the 2016 roster instead of waiting around for the overhauled system to produce another pre-arb-but-great reliever. The Smith fans would say that you can’t argue that reliever salaries are skyrocketing, or that relievers are increasingly seen as critical to team success, to justify the Cishek signing *AND* accepting “just” Miley and Aro in the Smith deal. The M’s would say that they used Smith to get something the team needed – starting pitching that would help the club in 2016 – and then replaced the bullpen hole by dumpster diving for an undervaled Cishek. Others counter that you can’t get a ton of credit for filling a hole that you’d just made after not accurately assessing how the league currently values relief pitchers.
That’s a lot of hypothetical arguing, and as is my wont, I’m not really interested in weighing in on one side or the other. If you read this blog at all over the past few years, you know I’m perhaps irrationally exuberant about Carson Smith, so I’ll say it was not a great imaginative leap for me to write from that side. But there’s definitely a case to be made that given salary inflation in general, a cheap starter and a cheap-but-great bullpen arm is worth more than a really cheap Carson Smith, some implied cash savings, and shopping for starters on the open market.*
Thus far, we’ve talked a lot about value and markets and not much about Steve Cishek, new Seattle Mariner. Let’s rectify that. Cishek has a very low arm angle, like O’Day’s, and has a similar pitch-mix: he throws 50-55% fastballs (overwhelmingly sinkers) and 45-50% sliders, with a handful of change-ups (a splitter, in Cishek’s case) mixed in. Like O’Day, Cishek throws his slider to lefties a lot, and like O’Day, he likes to keep his slider *away* from lefties. This has a couple of ramifications. For one, it allows him to grab some strikes-looking, and for another, it gets lefties to hit soft fly balls. Why? Because fly balls are more likely than GBs to be hit to the opposite field. The development of this backdoor slider resulted in both better K rates AND better results overall against lefties. Combined with a slight change in his approach with his fastball, and Cishek’s GB rate plunged in 2014 while his K rate rose.
This is essentially Darren O’Day’s approach. Despite the low arm-angle and the flurry of sinkers, O’Day is a *flyball* pitcher. Like Cishek, lefties hit fewer grounders against him than righties, but neither hits that many. O’Day shows the sinker early, then gets whiffs with his (slow) four-seamer by throwing it later in the count after batters, especially lefties, have seen his slider and sinker. Cishek *used* to do this, but hasn’t thrown his four-seamer much at all in recent years. Brooks has his throwing zero in 2015, but that’s likely a classification error – but that only highlights that Cishek wasn’t able to make his pitches distinct last year. O’Day’s command allows him to post walk rates that are much better than league average, but even so, he’s careful with lefties – he doesn’t pound the zone against them as much as he invites them to swing at pitcher’s pitches, only some of which are strikes. That was Cishek’s plan in 2014, but it fell apart in 2015. Last year, he walked/plunked 18 lefties, while striking out…18. Coupled with the somewhat alarming velocity loss, there are very good reasons why Cishek was first traded for a minor prospect midway through the year and then available cheaply in December.
Still, Cishek had a four-year track record in Florida that compares well with plenty of good relievers. The year after that run wasn’t great, but even with the lower velo and control problems, Cishek missed some bats and kept the ball in the park. After moving to St. Louis, the BABIP pendulum swung all the way back. While BABIP luck isn’t how you make your case for a new signing, it at least suggests that his ability to avoid barrels wasn’t entirely lost. It’s not so much that his BABIP was low in St. Louis as that the absurdly high rates in Miami weren’t indicative of his true talent.
Another player who has a similar approach and is even signed to a similar deal is the Astros’ Pat Neshek. Neshek has a funky arm angle, but, like Carson Smith, ends up delivering the ball higher than O’Day/Cishek. Neshek is something of an extreme fly-ball guy, and again, lefties hit more flies than righties do thanks in part to his delivery hiding the ball. He throws 90-91, or in between O’Day and Cishek, and his command is probably the best of the three. Late last year, Neshek started losing some velocity and he ended up scuffling badly down the stretch, though of course his Astros teammates were all playing poorly. Still, Neshek’s an example of someone with the same basic repertoire who’s able to succeed (mostly) at 90-91.
If he isn’t hurt, Cishek should add some value to the M’s pen, though exactly how much is hard to say. Steamer isn’t bullish, with a FIP/ERA in the David Rollins range. If he’s all the way back to his 2013-14 peak, he could add 1.5-2 FDP-WAR over the course of the deal. At this point, the M’s are handing him the closer’s job, so if he *is* effective, the innings he pitches figure to be pretty important ones.
* Yes, I realize the omission here is any discussion of Roenis Elias. If you think he’s got some value (and some of the projection systems do), then the M’s POV is harder to understand. If you don’t, and there are likewise many who’d agree with you, he’s simply not good enough to change the shopping list very much.
** This is just a random aside, and I thank you for reading all of this just to get some random thoughts that don’t quite fit the rest of the post, but here goes: the wisdom of the “buy low” FA signing seems well-studied and supported and all of that, but how often do RELIEVERS lose it for a year or two and then bounce back? I mean, we say relievers are volatile, but are we really saying that a successful reliever has only one way to go? Look at a list of great relievers from 2012-2014, and sure, many of them fared poorly in 2015, though injuries had a lot to do with that. Who are the guys who were great, then struggled, then turned good again? I hate to bring it up, but does Fernando Rodney count? I think Joaquin Benoit probably does, as does Joe Nathan. Mark Melancon’s 2011-2012-2013 looks purely, violently volatile, but while Glen Perkins’ 2015 was better than his 2014, but it wasn’t anything like his 2012-13. Anyone know of any studies on this?