Relievers Getting Paid

Yesterday, Todd Jones signed a one-year $7 million deal with the Detroit Tigers. This is nearly double what I have him valued at for 2007, $3.59 million. This follows J.C. Romero’s deal, which I also think was a bit high. I’m open to the possibility that I am undervaluing relievers—I discuss this briefly in my book—but I don’t think I’m off by this much.

I don’t like giving big contracts to relievers. They pitch very few innings and there is always the possibility of injury. I prefer the shotgun approach: bring in a bunch of relievers for cheap and find a few that are at the top of their games. Use free agents and farmhands who might not be ready to start. Sometimes this doesn’t work, but you diversify your risk and the payoff of having someone blossom who hasn’t become a free agent yet is significant.

Addendum: Tom Verducci at explains Kevin Tower’s philosophy for finding relievers, and I like it.

The risk of sinking $4 million a year over multiple years for a pitcher in his mid-30s (Romero turns 32, 33 and 34 over the contract) who doesn’t start, win or close games, and has high mileage on his odometer is one you won’t find Kevin Towers taking. The San Diego general manager is the industry expert at building a bullpen on the cheap.

Towers’ philosophy is that relief performance tends to be fungible, and buying free agent relievers — who tend to be older and overworked by the time they get to the market — is the definition of buying a stock too high before the regression hits. Think Danys Baez, Arthur Rhodes, Kyle Farnsworth, Tom Gordon and Hector Carrasco.

“Free agent relief shopping is dangerous,” Towers says.

3 Responses “Relievers Getting Paid”

  1. Marc Schneider says:


    Your theory makes sense in the abstract but the problem is that, while relivers might be fungible in some sense, there is a significant risk to counting on finding diamonds in the rough. If none of the pitchers work out, your whole season can be shot. The process worked well for the Braves for several years but when it didn’t work, the Braves were screwed because in today’s game you can’t win with a bad bullpen as the Braves found out. So, to me, if you are a team that expects to contend, it makes some sense to overpay for some reliability in the bullpen. (Whether it makes sense to overpay for Todd Jones, specifically, is another issue.)

  2. David says:

    I’d agree that that can work for most of your bullpen, and in some cases all of your bullpen – Takashi Saito was a NRI for the Dodgers in 2006. I think the reason teams still value closers like they do is because the marginal cost of uncertainty in the 9th inning is greater since that allows for the fewest run-scoring oppurtunities, especially on the road. Because of the sudden-death potential here, GMs like to follow Rawles here – behind the veil of ignorance, you would rather have a proven closer to start with than risk not having a reliable 9th inning reliever.

    That still doesnt’ justify paying Todd Jones that much, though. When I’m behind the veil of ignorance, I want a closer who can put up K/BB numbers as close to Ben Sheets 2006 as possible.

  3. Jonathan says:

    In my mind, the Phillies’ risk with Romero is much greater than the Tigers’ with respect to Jones: assuming, for a moment, that Jones’ “true value” next year is $3.59 million, the Tigers are mortgaging around ~$4 million in expecting him to reach that level — and there is the added benefit of their retaining him before any other team has a chance to make an offer. Since Jones has put up 116 and 107 ERA+ numbers for the Tigers in the past two years, in 64.0 and 61.3 IP respectively, this gamble doesn’t look horrible for a team that was recently informed it would be without Joel Zumaya in 2008 — especially with a player they know.

    The Phillies, meanwhile, are mortgaging $1.5 million in 2008, 2009, and 2010 in the belief that (if we use JC’s numbers) Romero will maintain his 2007 value; Romero, it should be noted, posted 68 and 243 ERA+ numbers the past two years, in 48.3 and 56.3 IP respectively. Of course, the Phillies too were able to retain him before other teams could bid for his services.

    The difference? Well, besides the fact that Romero has been more erratic lately, and slightly worse for his career relative to Jones, the nature of each team’s risk/reward incentives are very distinct. The Tigers filled a recently-emergent need for 2008, but 2008 only; moreover, they payed a few million extra to secure a reliever who pitched fairly well with them in the prior season, instead of venturing in the free agent wasteland of more-uncertain reclamation projects and more-uncertain multi-year deals. The Phillies, on the other hand, are making a slightly cheaper bet for three consecutive years – itself an inherently riskier proposition – and seem to reap no reward for signing Romero early: Jones may have received two years as a free agent, or taken less money to play for the Braves (as was reported), while Romero was not likely to get any more years or guaranteed money elsewhere.

    Naturally, with both these clubs being reported as having around $10 million left to add to their payroll, and needing starters, the extra millions could have been spent to fortify the clubs’ respective rotations; but, once again, the Tigers were in a better position to make such a gamble (in terms of their overall pitching needs), while the Phillies have sunk $12 million into Romero and still need two starters for next season.