K-Rod signs with Mets
Last week, we talked about how the market for relief pitchers appeared to be crashing. Today, Francisco Rodriguez and Kerry Wood became the latest to feel the contraction in bullpen spending. K-Rod is signing a contract with the Mets that will guarantee him $37 million over three years, while Wood is close to signing a two year deal with the Indians, who aren’t exactly known for spending significant portions of cash on free agents.
Rodriguez gets less money guaranteed than Francisco Cordero got last year, and I don’t think too many people believe that Cordero is the superior reliever of the two. Perhaps more teams were scared off by the decrease in velocity than K-Rod’s agent expected, but even still, this is half of his rumored 5 years/$75 million asking price heading into the winter. At this point, I don’t think there can be any question that MLB teams are valuing relievers significantly differently than they have in the past.
However, that doesn’t affect Rodriguez’s actual value, so let’s take a look at what the Mets should expect from him in 2009 to 2011.
Marcel projects Rodriguez to throw 66 IP with a 3.40 FIP, but it doesn’t know that he’s switching leagues, so we need to knock that down a bit. We’ll call it 3.25 (the adjustment for a closer is less because they never face pitchers), which of course is very good for a reliever. If we set replacement level for an NL reliever at 4.75, then Rodriguez would be 1.5 runs per nine innings above replacement, which translates into +11 runs per 66 innings.
However, we have to add in a leverage factor for late inning relievers, because the runs they allow are more valuable in terms of wins and losses than a normal reliever. 1.8 is the average leverage factor or a closer, so 11 * 1.8 = 19.8. We’d say that Rodriguez, as a closer, is worth about +2 wins more than a replacement reliever.
At an annual average salary of $12 million, that means New York is paying about $6 million per win. For most teams, that just not feasible, but as we saw, the Mets payroll allows them to have an average of nearly $3.4 million per win across the whole payroll. Considering that they have significantly underpriced players such as Jose Reyes, David Wright, and Carlos Beltran (his contract is a bargain compared to how good he really is), the Mets have the luxury of being able to overpay and still come in under budget while contending.
On it’s own, the contract isn’t great, and he’ll be hard to trade if they ever decide they don’t want him anymore. But in the specific circumstances that the Mets are in, needing a closer and having significant financial resources that most teams don’t, this deal makes sense for them. Not many teams are in the position to pay $6 million per win – the Mets are.
I think this was a move Minaya had to make. After the September struggles of the last two seasons and the abysmal closer carousel of 2008, he would have been eaten alive if he didn’t sign a top FA closer.
Also, the fact that the Mets are a borderline playoff team makes this signing even better. Even if they technically are overpaying by just a bit, I like the deal for them.
I’ve just begun using pLI for relievers as I learn how to “homebrew” this stuff. I wsa using 2 for closers and 1.5 for setup guys. Are 1.8 and 1.4 more accurate, you think?
If so, then the Mets overpaid even more for F-Rod than I thought. If he’s a +2 WAR guy who doesn’t decilne, he should get 3/33, but I don’t think he’s quite a +2 guy, at least not according to Marcel, especially with an expected LI of 1.8.
Wrong wrong wrong! See my NY Times column at http://www.nytimes.com/2008/12/07/sports/baseball/07score.html. You are not factoring in chaining–the fact that the replacement reliever doesn’t pitch the closer’s innings; he pitches the mopup man’s innings, and everyone gets bumped a role. See post #55 at http://www.baseballthinkfactory.org/files/newsstand/discussion/mets_close_to_3_year_deal_for_k_rod/ for a work-through example.
While this may be overstating the pLI of a closer a little bit since he’ll appear in some non-save situations, the average pLI for a player who either recorded a save or blew a save in a game was 2.27. If the Mets use him as they should, the cost could be more like $5 million per win, but I don’t think it gets a whole lot lower than that.
Well one thing (and I don’t know how much or how little this really affects overall run prevention) is that when you sign a closer, the guy who would have been a closer becomes a set up guy, the set up guy becomes a 6th/7th inning guy, and so on. It sometimes will allow to use good lefty as a loogy instead of a set up guy. In general it probably decreases the pLi a little bit for all of the other relievers. It’s probably only a very small amount of actualy value, but I definitely think it’s something that GMs think about when making deals like this.
Don’t worry, Dan, I read the NYT piece already – I just disagree with your way of doing things when it comes to valuing closers. Doesn’t mean you don’t do really good work on other stuff, though.
I mean, what he said.
Would you care to clarify what you believe the flaw in the logic to be, Dave? (That said, we’re arguing about 0.35 wins here…the error bars on our projections are what three times that large?)
I don’t believe that teams have bullpen spots 1-6 that players will move between depending on their teammates. In general, they all have a relief ace (or someone they plug into that hole, anyway), right-handed and left-handed setup guys, a middle reliever who with a rubber arm, and a long reliever/spot starter. Most teams now use their 11th pitcher spot to roster a LOOGY, and the teams who carry 12 pitchers will go with two middle guys if they can’t find a rubber arm type.
In general, teams do not actually bump guys from relief ace down to setup guy, and then setup guy down to middle reliever, and then middle reliever down to mopup guy, when they acquire/develop a relief ace. The modern bullpen has distinct roles, and most players get pigeonholed within a role until they perform to a point that they get moved out of that role.
There just aren’t teams with three relief ace types where one is being under-leveraged because of his teammates.
In a bullpen, every pitcher tends to rise to his level of incompetence.
So what ARE you saying happens when a closer goes down? The right-handed setup guy you mention becomes the new closer, the middle reliever with a rubber arm becomes the new right-handed setup man, and the actual replacement reliever become the new middle reliever (or half of one). That’s at least three “chain”-worthy steps…no?
Dan – I think the flaw in your logic in this particular case is that every reliever on the Mets was varying degrees of suck last year. Its not as if they had some second ace waiting in the wings to replace wagner causing the chaining effect you’re talking about. In fact, they went out and got Ayala and inserted him as their closer. So in this particular case, the Mets are replacing a sub-replacement reliever (Ayala) with Rodriguez. Granted, this is overly-simplistic since Wagner pitcher the first four months, and its the Mets own fault for trying Ayala in the ninth inning, but the fact is, they Mets bullpen (pre-K-Rod) had 0 pitchers who could be expected to strike anyone out (aside from Aaron “3-run homer” Heilman”) or who could get both lefties and righties out. I’m not saying that K-Rod is the salve that will suddenly transform the entire relief corps, but I don’t think the chaining analogy works in this particular case, if ever.
Dan – I also wonder if chaining is appropriate given the context of it being early in the off-season. The Mets were going to acquire a “proven” closer. Using chaining to value any of them doesn’t make much sense as they would all have had identical chaining effects (assuming they would have all pitched an equivalent number of innings for the Mets given the same number of “closer-worthy” situations). Now if the Mets make a late move or a trade during the season and the rest of the staff was set in stone, then I think chaining might be a worthwhile argument to make.
What we should be evaluating (and I think what Dave is doing) is measuring how did the Mets do in the contract they are giving K-Rod versus the established market for closers. Determining his value versus what the Mets’ alternatives are/were is where the action is.
It’s also interesting that until the K-Rod deal, every FA signing so far this winter seemed to be for significantly lower than $5M / WAR.
So what ARE you saying happens when a closer goes down?
In the off-season? They get a new closer.
If this was July and the Mets were currently using Aaron Heilman as their closer, I’d agree with you. It’s not. The Mets were always going to acquire a proven closer this winter. There’s no bullpen chaining going on in Queens.
Hey Dave,
In your article you mention we need to “knock down” K-Rod’s FIP “because he’s switching leagues.” I’d agree with this in theory since NL pitchers get to face their pitching counterparts a few times a game, but when does a closer ever face a pitcher? Like I said, in theory NL pitchers do. But, the reality is closers would never face a pitcher, but rather a pinch hitter.
EDIT: whoops… I didn’t even keep reading the article. I see you accounted for closers never facing pitchers. But why would you knock down the FIP at all then?
He knocks down the FIP because the National league is poorer offensively without counting the pitchers.
theres also the DH vs PH thing. Having a DH in the lineup is much more conducive to offense because DHs are paid to hit whereas PH are more often role players who are better than the worst hitter in the lineup (and pitcher) but not good enough to get an everyday gig
So it looks like the K-Rod luck train is finally starting to de-rail.
Since June 18th (the day he blew a save vs. Baltimore):
19 G, 7 SV, 1-2, 8.15 ERA, .274/.398/.507, 1.98 WHIP, 1.067 K/BB, 7.64 BB/9.
SSS alert, sure, but his numbers have screamed regression for a long time now, and now he’s been rocking a WHIP nearly 2x his K/BB, and he’s currently being hit like he’s facing A-Rod every batter faced. Awesome $37,000,000, Mets.
And evidently I’m the only man who cares.
It’s ok, I’m still reading what you right, Joe.
Wow, obvioulsy I mean “write”, not “right”. Cringe-worthy error there.