Pondering the Coke Deal

facebooktwitterreddit

Now that the terms of Phil Coke’s contract for 2014 have come out, the deal is if anything more puzzling than before. We expected that the Tigers had signed Coke at a discount compared to what he would have received in arbitration (the minimum increase over his 2013 salary) and that did turn out to be the case. However, the discount is not large – a pay cut of $200,000 which could drop to $50,000 if Coke meets all of his performance benchmarks.

It’s simply difficult to understand this deal in the context of the Tigers other two big bullpen changes so far this offseason: declining Jose Veras’ option (for a little over $4 million) and letting Darin Downs go on waivers to the Houston Astros. Veras was not worth $4 million and a roster spot. Darin Downs was not worth $500K and a roster spot. Coke IS worth $2 million and a roster spot.

Oct 15, 2013; Detroit, MI, USA; Detroit Tigers relief pitcher Phil Coke (40) comes into pitch during the ninth inning against the Boston Red Sox in game three of the American League Championship Series baseball game at Comerica Park. Mandatory Credit: Andrew Weber-USA TODAY Sports

Tigers fans have largely – I would say – tired of Coke and expected that his time with Detroit had come to an end. $2 million is a big salary for most of us, but it isn’t a particularly expensive price to pay for a solid late-inning reliever. What I’m saying is: no one figured Coke to get the boot because he was too expensive or because the Tigers needed to free up cash, it was more a question of whether he was worth a roster spot and that I must say is debatable.

Coke’s 2013 was pretty atrocious, there is no way to sugar coat it. The 5.40 ERA jumps right out at you – and that came in spite of the fact that he was used in matchups more than ever before in 2013. He was bad against lefties and bad against righties, bad at home and bad on the road, bad early in the season and bad late, even bad in the postseason (and with in a box with a fox). It’s also worth noting that Coke’s career numbers – the “track record” as Leyland would have said – aren’t spectacular either: a 4.20 ERA and a strikeout rate of only 7 per 9 innings (good for a starter, not so good for a reliever).

The Tigers clearly like Phil Coke and more importantly like Phil Coke’s stuff. They must believe that 2013 was an aberration and they must believe in FIP over ERA (since Coke’s continually high batting averages on balls in play make the two numbers diverge). Projection systems would more or less agree – Steamer predicts a 3.93 ERA for Coke next year, with a BABIP down to just a hair over league average and a walk rate back closer to Coke’s career average. It was, after all, mainly his inability to locate that made Coke such a poor performer in 2013. If you believe what Steamer says – and I’m not sure that I do – $2 million is not a bad rate to get that guy, at least if you don’t see any low-cost internal options to use instead. But Coke comes with risk attached, that’s a fairly optimistic projection and he could be worse than in 2013 and not better. Relievers are always volatile and while that means that you often see guys crash and burn only to be resurrected as elite relievers (like Javier Lopez among others) you also see a lot of guys that just fall out of the bottom of the league.

Lets compare Downs (who wasn’t worth a roster spot) and Coke (who was) for a moment: Downs also had a poor 2013, with a 4.84 ERA that got him booted back down to Toledo. Downs also has a relatively high career ERA now (4.34) with a big split between his ERA and FIP due to the BABIP. By FIP and by peripherals, Downs looks like the vastly superior pitcher to Phil Coke in 2013 – downs only real problems were the high BABIP and extraordinarily large L/R splits. Downs also has much less of a “track record” than Coke, his career “mediocrity” is solely a function of a good 2012 and a bad 2013 (by BABIP and ERA) rather than the sort of sustained below-average-ness that Coke has provided. If you look at SIERA – which predicts an ERA based on things like how many hits of what kinds the guy allowed – Downs looks like a very good pitcher, Coke a pretty poor one. Steamer (and again, I don’t think keeping Coke makes any sense if you don’t follow the same personnel evaluation logic as a model like Steamer) predicts a 3.90 ERA for Downs next year with very similar peripherals to what it predicts for Coke. Basically, Steamer thinks Coke and Downs will be basically the same pitcher next year – as far as numbers go.

It’s easy to guess at why the Tigers would want to keep Coke but let Veras go – Veras was a little more expensive and he was a righty. They may figure they have more candidates in house to replace Veras and that there are more free agent options that could replace Veras at a similar price. They may be right. They may feel that there aren’t a lot of lefties out there to pursue, particularly now that Javier Lopez is off the market. They may also have made a final decision to move Smyly into the rotation, which would mean the team needs one more good lefty than they would have otherwise. In short, Coke’s “lefty-ness” may be all that matters in that comparison.

I’m still puzzled as to why a team that needed a lefty would have let Downs go – and especially why they would have preferred to spend an extra $1.5 million to have Coke instead of Downs. I can really only think of two possible reasons: 1. All Dombrowski cares about is how hard the guy throws. Coke throws 3 MPH harder than Downs, though without the right movement or location for that fastball to actually be more effective than Downs’ fastball. Downs relies on a consistently filthy slider to finish hitters, Coke’s slider is not nearly so reliable. In terms of “stuff” Downs is the winner, in terms of velocity it’s Coke. This would mean that DD always intended to keep Coke, never though about keeping the other two. 2. There has been a major realignment in strategy since they let Downs and Veras go. Because of the way arbitration works, the Tigers had longer to think about Coke than Veras (Downs they just didn’t want). They may have come to the conclusion that there isn’t going to be the glut (with associated low prices) of free agent relievers that they had expected. They may have only recently decided that trading a starter was a definite rather than a slim possibility. It could be that Kenny Faulk is not pitching well in the AFL and that Matt Hoffman did not pitch well in Venezuela (and could be hurt, news is scarce) making Tigers brass leery of giving a job to one of those two in spring training. There may also be some influence of the new manager and his new staff on these decisions, though it’s hard to imagine Ausmus stepping in and telling DD that he’s got to have Coke.

The Tigers probably wouldn’t be able to get a replacement for Coke on the free agent market with more talent and fewer question marks for less money. BUT, they could definitely have gotten guys with more talent and fewer question marks without paying absurdly high prices (maybe $3 or $4 million per season). They could have gotten guys with a similar – maybe slightly better – talent/risk profile for the same or less. Oliver Perez? Scott Downs? Mike Gonzalez? Matt Thornton? These guys are out there… But they elected to go with Coke. After all this pondering, I still don’t get it.