I was going to post something about relievers this week, but Omar Minaya presented too rich of an opportunity to begin applying some of the tools from my post on pitcher value.Yes, it's another blog post on the wonders of the Mets' signing of Oliver Perez.

You've probably read that Oliver Perez and the Mets have reportedly agreed upon a 3 year contract worth $36 Million. Perez is only 27. He had a very good season in 2004, and a pretty good 2007. He's been... less than awesome pretty much every other season. But that doesn't matter now, not exactly. We want to get an idea of good is he going to be going forward, and what is he worth, really?

I'm not going to do my own projection, that's not what this is about. I'm going to use his CHONE projection and also his Stat Corner tRA* numbers. We're just going to take those numbers use them to project Perez's performance and compare with what the Mets gave him with the contract.

Earlier this offseason, I put together a salary chart to see what teams are paying for in their free agent contracts. It was meant to be used with position player free agents who have a different decline curve than a slightly younger pitcher like Perez, but you can still see what the Mets are paying for per year. Assuming that each win above replacement (WAR)  is worth $4.8 this offseason, $12M a year (at least in the first year) for a player implies a player worth about 2.5 wins above replacement. Perez's Fangraphs page says he's only reached 2.5 WAR once in his career to date, back in 2004 with the PIrates. But we want to figure this out ourselves.

[NB: I'm leaving aside the issues with figuring runs-to-wins conversions for pitchers, as it's a bit more complicated and when projecting forward even moreso. I'll just use a generic 10:1 runs:win ratio.]

Remember from the pitcher value post that player value = rate over replacement x playing time. For starting pitchers, this is

((lgavgRate*1.28)-playerRate)*(IP/9.0)

Let's use FIP first. According to Fangraphs, CHONE has Perez projected for a 4.97 FIP-ERA over 166 innings). Using a simple 5-4-3 weighted average, let's project a 4.39 2009 league ERA. So we have

((4.39*1.28-4.97)*(166/9) = ~12 runs above replacement, or about 1.2 WAR. That doesn't even come close to what the Mets are paying.

Now the projection systems (at least CHONE, from what I've read from the creator) doesn't mean the playing time projectio as a "prediction." It's just to set up the rate. Let's give Perez 200 innings. Probably too generous, but he's a young guy, so maybe... Nope. 14.4 RAR, or (maybe) 1.5 WAR. Again, not close. Even before the seeming crash in the free agent market, a 1.5 WAR player, even assuming he doesn't decline at all, is worth a $25M contract. And that includes a lot of optimistic assumptions.

Maybe Perez is getting shorted on his WAR by using an ERA-scaled stat. Let's use Stat Corner's tRA. There isn't a projection system that uses tRA yet, so we'll do a 5-3-2 weighted average of Perez's tRA* (tRA regressed to yearly league-average rates for home runs, etc.), and we get a 4.97 tRA projection for Perez in 2009. We have a weighted average lgRA of 4.76 for the NL. Over 200 innings, Perez is about 25 RAR. So maybe the deal is worth it. Notice, though, that this doesn't come from a real projection system. Moreover, Perez hasn't pitched 200 innings in his career, and the only system that projects him to do so in 2009 is the notoriously optimistic Bill James projections.

Going to CHONE's expanded page for Perez (which uses different dollar values than I did here), for the next three years Perez's projected WARs are 1.6, 1.3, and 1.1, for a total dollar value of $19.7M over the next three years. My feelings are about the same. At most, for three years Perez should have got $27M, and that's being quite optimistic and pushing it.

So only one very rough projection shows Perez to be close to worth the money, but it assumes him pitching more innings than he ever has. Everything else shows the Mets to be way off.

It's a bad contract, most likely. Will it kill the Mets? Probably not. They have a lot of money, and it can't be any worse than the Luis Castillo deal, although the Citigroup rumblings aren't good.

The most problematic Mets contract, from the looks of things, is Omar Minaya's.

[NB: I didn't realize it until after I published this article, but Driveline's own David Golebiewski came to very similar conclusions as to Perez's value a couple months back in this piece. Nice to see the resutls dovetailling. Unless you're a Mets fan, of course.]