Choose Your Own $38 Million Starting Pitcher

Yesterday, Tyler Chatwood signed with the Cubs for $38 million over three years. Given his velocity and run prevention at altitude, there’s a reasonable case to be made that Chatwood comes with enough upside to make this a very intriguing bet by the Cubs.

But while Chatwood remains interesting, one thing he can’t be described as is durable. He’s had Tommy John surgery twice (once in high school), missed nearly all of the 2014 and 2015 seasons, and has never thrown more than 158 innings in a season. The history of guys who have already had Tommy John revision surgeries is not very good, and quite simply, Chatwood’s health is a legitimate question mark. While the Cubs bought some real upside here, there is also a non-zero chance that they just spent $38 million to watch Chatwood spend most of the next three years rehabbing his elbow.

So, if we were plotting all pitchers on a risk/reward graph, Chatwood would be about as far from the middle of the graph as a point gets. Interestingly, though, the market also recently decided that the guy at the very opposite end of this spectrum is also worth $38 million over three years.

At the end of August, the Mariners acquired Mike Leake from the Cardinals. At that point, Leake had about $55 million left on his deal, but the Cardinals kept $17 million of his remaining salary on their books, so the Mariners took $38 million of Leake’s deal, covering the last month of 2017 through the 2020 season. There was also a low-level prospect in the deal, but that was probably to offset the fact that the Cardinals sent the Mariners some international bonus money in the trade, as the team has been collecting money for Shohei Otani for a while now.

The Mariners effectively got Leake for 3/$38M without surrendering any real talent, much like if they signed him to that price as a free agent. And while 3/$38M got the Cubs some upside and risk, Leake brought the Mariners some fairly predictable stability. Leake has made 30 or more starts for six straight years. He has never had any kind of arm problem, only ending up on the disabled list for short stints due to a hamstring strain and shingles.

Like Chatwood, Leake is a groundball guy who doesn’t get that many strikeouts, though he gets even fewer of both. He compensates by throwing a ton of strikes, however, something Chatwood has not done regularly. After a velocity spike in 2015, Leake went back to sitting 91 last year, so he’s more of a traditional pitch-to-contact guy, while Chatwood is more of a guy who hasn’t got that many strikeouts but looks like maybe he could if he got away from Denver.

And yet, despite their differences, they’ve ended up at somewhat similar results overall.

Chatwood has outperformed his FIP/xFIP, in part because he’s run low-for-Coors BABIPs during his time with the Rockies. Leake has gone the other way, running an ERA a bit lower than his FIP but higher than his xFIP, as he’s had a sustained home run problem that isn’t regressing to the mean. So, effectively, Leake has been a bit better when the ball isn’t put in play, thanks to a walk rate that is half of Chatwood’s, but Chatwood has been a bit better on contact, especially once you adjust for their home parks.

That said, we can strip out some of the effects of Coors — not all, since the altitude also affects pitch movement and selection — if we just look at exit velocity and launch angle data. And MLB uses those calculations, then adjusts for walks and strikeouts, to come up with their xwOBA calculation, so it rewards guys who legitimately are inducing weak contact. And here’s how these two compare over the last two years, since Chatwood has returned from TJ surgery.

Unsurprisingly, the guy who pitches half his games in Coors Field has had worse results than his batted ball contact would suggest, but because he’s been a bit fortunate on getting balls in play turned into outs on the road, it’s not as dramatic a gap as one might expect. By both wOBA and xwOBA, Chatwood and Leake have been very similar the last few years.

But, teams aren’t paying for what players have done previously, and there are some more factors to adjust for when looking forward. For one, Leake is two years older, having just turned 30, while Chatowod turns 28 next week. Chatwood’s additiona velocity also gives him a higher floor to age from, as he can lose a few ticks and perhaps still be valuable, while Leake is closer to that line where a bit of velocity loss might turn him into a fringe starter.

Additionally, if Chatwood is healthy in October, he’s probably easier to deploy than Leake. Even if he doesn’t crack the Cubs playoff rotation, a guy who sits 95 and gets groundballs as a starter is the kind of arm a manager would have no problem trusting in high-leverage relief innings, so Chatwood could reasonably be expected to transition into something like the Kenta Maeda role if need be. Given the cost of relief pitching right now, having a guy could probably be an effective reliever if he doesn’t start in October gives the Cubs multiple chances to extract value from him in the postseason.

With Leake, if he’s not your #4 postseason starter, he might not even make the roster. He doesn’t have the kind of stuff that one would expect to play way up in the postseason, and his throw-strikes mindset isn’t what teams are looking for in high leverage, must-prevent-runs scenarios. And with teams refusing to let their starters dig big holes in the postseason anymore, even the idea of an October long reliever is now somewhat outdated, so if Leake doesn’t crack your playoff rotation to make one start per series, he might not help you in the postseason.

Leake and Chatwood are very different pitchers with very different repertoires and very different medicals. In both cases, the market valued them at 3/$38M. So, let’s put this to a poll. For $38 million, would you rather have the safe strike thrower or the risky stuff guy who might benefit from leaving Coors Field?