Cole Hamels: How Much Is Too Much?

We've been pretty close to unanimous around here in believing that the Phillies should do whatever it takes to persuade Cole Hamels to sign a contract extension -- myself included. But lately I've been starting to wonder if we ought to be a bit more circumspect about this. Anyone can be overpaid, and some of the numbers that have been thrown around in connection with Hamels might be edging rather close to the line.

This is not to suggest that Hamels is anything but an outstanding pitcher. Last year, he was 7th in the NL in ERA, 9th in FIP, 7th in SIERA, and 5th in xFIP (of 75, min. 100 IP). He just had his 28th birthday last week. And he's suffered no major injuries in his big league career, just some nagging ones here and there. So he's clearly worth a lot of money. On top of all that, if there's a part of you that's rooting for him to get his payday because of how he's been mistreated or underappreciated by the fans at times in his career, then that's also very understandable. I don't think Hamels is underappreciated today, but he was certainly the recipient of a lot of unjust abuse back in 2009, and there would be something poetic about stamping out the haters one last time.

Nevertheless, here are four points that should perhaps give us pause.

1. Hamels has never broken 5 fWAR in any year of his career.

Here are Hamels' fWARs since 2007, his first full season in the majors: 3.8, 4.4, 3.6, 3.7, and 4.9. That's an average of 4.1 per year, which would be worth a little over $18 million in 2011 dollars, according to Fangraphs. The career-high 4.9 that he posted last year at age 27 was worth $22.1 million.

That's nothing to sneeze at. If you're worth $18 million, then you're pretty darn good. But the numbers I've been hearing for Hamels have been $20 million per year, minimum, and probably higher than that. Will that really be worth it? You've got to at least feel some trepidation about that question.

(WAR is somewhat problematic for pitchers because it's derived from FIP, and over his career, Cole has generally posted better xFIPs than FIPs. That wasn't the case in 2011 though, when his FIP and xFIP were only 0.03 apart, so it's still legitimate to say that he's never broken 5.)

I think we can surmise that if Hamels does sign an extension, his per-year salary will be in the same ballpark as his current per-year value. That means that the wisdom of the extension from the organization's perspective will be dependent on one question: Which risk is greater? (A) The upside risk that Hamels will continue to improve and/or that there will be substantial salary inflation. Or (B) The downside risk that Hamels will fall off or get hurt.

And I suspect that the latter is greater. See points 2 and 3 below.

2. Hamels probably doesn't have that much "upside" left because he isn't that young anymore.

I think we sometimes feel like Cole is younger than he really is just because most of his teammates are older in comparison. But his DOB is 12-27-83, meaning he's young but not that young. He's probably right in the middle of his prime. So while he isn't likely to go into decline for a long while yet, he also isn't terribly likely to improve much more. Chances are that if he hasn't reached his peak, he isn't far from it.

Note that I said "likely." Clearly, strange things happen in this game from time to time. When Cliff Lee was the same age, he'd barely accumulated 9.0 fWAR over about four seasons worth of major league starts. Then, the next year, he won the AL CYA and he hasn't looked back since. Still, that kind of thing is unusual. It's encouraging that Hamels had his best season ever last year at age 27, and it's possible that he'll turn into a 6 or 7 WAR pitcher at age 28 or 29 or 30, but I don't think you'd want to bet a ton of money on that.

Again, there's nothing wrong with what Hamels already is. If he's "only" likely to stay around his current level for the next few years, then he's likely to be an excellent pitcher for the next few years. But if his average salary is going to be commensurate with his current level, and if his potential "upside" of improving beyond that level is low, then his expected value will fall short of the cost of the contract unless his "downside" risk is equally low -- and I don't think it is (see point 3).

3. Hamels' downside risk is fairly low but it isn't insignificant.

In comparing Hamels' upside risk to his downside risk, we need to look at two things: the probability of a step up or down, and the potential magnitude of a step up or down. I think the probability of Hamels declining or getting hurt is smaller than the probability of further improvement -- but not by that much. And the potential magnitude of a step down is way bigger than the potential magnitude of a step up.

It isn't a new insight to point out that pitchers are inherently risky, even those who have no injury histories. Recall that in January 2007, Freddy Garcia was thought to be one of the most durable pitchers in all of MLB, and you know what happened after that. Even very durable pitchers are only durable until they aren't. And in any event, Hamels does have an injury history of sorts. Not in the majors, but when he was a prospect, he had lots of yellow flags around him. In high school, he broke his pitching arm and slipped in the draft because of it. The injury probably happened in a game of street football, but he was unaware of it until his humerus snapped on the mound while he was throwing a pitch, a la Tom Browning or Tony Saunders. Then in the minors, the Phillies' developmental staff (appropriately) treated him with kid gloves, and he missed a lot of games with injuries. One of those injuries was fluky, as it was the result of a bar fight, but only one. In 2004, he was limited to 16.0 IP all year because of left elbow tendinitis. I realize that this history is ancient now, and maybe it's 98% irrelevant today. But I don't think it's 100% irrelevant.

Also, I note that Hamels' top two comps on Baseball-Reference are John Smiley and Greg Swindell. Those are two names I remember very clearly because they were both free agents in the 1992-1993 offseason, and the Phillies' front office got torched in the media for failing to go after either of them (or Doug Drabek, the other big FA pitcher on the market that year). Instead, the Phillies made a much less ambitious move by acquiring Danny Jackson, a fourth starter, via trade. As it turned out, Jackson had an okay year and the Phillies won the 1993 pennant. Meanwhile, Smiley and Swindell both fell off a cliff and performed horribly with their new teams. Smiley later bounced back and posted very good seasons in 1995 and 1996, but he suffered a career-ending injury (a broken humerus, actually) in late 1997 at age 33. Swindell was reincarnated a few years later as an effective relief pitcher, but he never had success as a starter again.

No one puts a great deal of stock in B-R comps and there are lots of specific reasons to discount them here. For one thing, the early '90s were a very different era. Swindell, in particular, was overworked like crazy when he was young, while the Phillies in general and Charlie Manuel in particular have been much more conscientious with Cole. (I sometimes wonder how much the current MLB-wide drop in offense and glut in pitching are attributable to teams having been much more cautious with their prospects' pitch counts in recent years.) Still, the comps are legit data points, and worth taking into account, at least a little bit.

In the final analysis, I doubt that Hamels is much more of a risk to fall apart than the average pitcher his age -- if he is, it's only by a slight margin. But even if the probability of a crash is low, the magnitude of the harm that it would inflict upon the team would be enormous -- it would basically mean $20 million per year down the tubes. Yes, the same could be said of any pitcher, but the beauty of the Roy Halladay contract and, to a lesser extent, the Cliff Lee deal is that this risk is priced into their terms. If Lee were to fall off a cliff (no pun intended), it would be equally catastrophic, but his contract is arguably worth the risk because if Lee stays healthy and doesn't decline much, then the Phillies get to pay only $24 million per year for a guy who's proven multiple times that he's able to perform at a level approaching $30 million per year. But that might not be the case with Hamels.

4. It might be more efficient to budget the money to be used on a hitter instead of a pitcher.

From where I stand, the Phillies' farm system seems to be a lot deeper in pitching prospects than position player prospects right now. If that's the case, it might make more sense to devote more money, at the margins, to hitters rather than pitchers.

Doing that might make more sense even if Hamels is better than the best position player alternative on a straight one-to-one comparison. Here's what I mean by that:

Let's say Hamels is willing to sign a $20 million/year deal, while his annual expected value over the next X number of years is $21 million. Let's also say that some hitter out there (we'll call him Zyan Rimmerman) is willing to sign for exactly the same contractual terms, but his expected value is only $19 million per year. If these are the only facts we have to go on, then you should clearly choose Hamels.

But now let's say that if Hamels leaves, the Phillies have a 75% chance of filling his vacancy with a homegrown prospect who will perform at a league-average level. Whereas, if they fail to sign Rimmerman, they have only a 25% chance of filling that void with a prospect who will be league-average. That changes the analysis because signing Hamels for $20 million = signing Hamels and a hitter for $20 million plus the cost of a second market-priced contract. Signing Rimmerman, meanwhile, is likely to cost the $20 million only, because you have the minimum-salary youngster to fill the pitching void. (Theoretically, you could sign Hamels and then try to trade your pitching prospect for a hitting prospect. But that's risky.)

This hypothetical scenario is admittedly oversimplistic, but still, of the sixteen Phillies prospects that John Sickels graded at C+ or better this offseason, twelve are pitchers, and only one of the four hitters (Sebastian Valle) played a full season of minor league ball in 2011. Devoting your next $20 million to a pitcher -- even an excellent one -- might not be the optimal strategy based on the odds.

* * * *

To be clear, I am not saying that the Phillies shouldn't sign Hamels to an extension. I'm just saying that it's a trickier situation than I, at least, originally thought. If $20 million is the minimum the organization needs to offer to even get into negotiations with Cole's agent (maybe it is, maybe it isn't), then they need to exercise some caution.