Statistically Justifying Long-term Reliever Contracts

[Erick]

It seems like the most difficult thing to statistically evaluate in baseball is the performance of a reliever.

Every offseason now, relievers are given long-term contracts (like the one we gave Bell) and fans/sabermetric analysts argue that it's not a wise use of resources.

However, it seems like just about every organization has done it, at one point or another. The only teams that won't do it are teams like the Rays and that's probably only because they have no other choice.

WAR seems like the common statistic to use now to evaluate signings, but some argue that it's a terrible statistic to use for relievers and that WPA is much better.

I'm intrigued to hear the opinions of most on here regarding reliever signings.

In terms of WPA, fangraphs had Heath Bell, for example, at a ~4.5 WPA in 2010. Obviously I'm not saying that's exactly who Heath Bell is, nor is this really about Heath Bell. However, I'd like to know what a reliever who puts up those #'s (hypothetically speaking) is worth in free agency, in your opinions.

It's been a topic that I've found really interesting especially the past couple of offseasons. It can't be as simplistic as "these teams are dumb; WAR clearly shows how overvalued they are." When the majority of GM's keep doing the same thing, I'm guessing they're evaluating this differently, no? It can't be a "hey let's do some shit because he did it, too." There has to be some sort of explanation for it.

[Bobbob1313]

I think a lot of it is conventional wisdom running amok inside of baseball without regards to whether it is sound policy, but at some point, market value is market value. Is Bell worth $9 mil a year compared to Josh Willingham getting a similar deal? Probably not. But compared to relievers of his caliber, he is.

There's also the question of whether all wins are created equal. Generally speaking, they are. However wins above a certain margin that push you into playoff contention are far more valuable, and should be weighted as such. Part of the problem with judging player contracts is it's so often done in a vacuum. But a team that needs a certain piece to get over the hump can be justified in spending to get that piece.

All that being said, I do believe relief pitchers are the most fungible asset in the game, so paying a premium for merely above average ones, like we did in the Maybin trade, is, I think a poor usage of resources. But for the guys who prove they can keep up top-notch production over a number of years, there's value in knowing what you're gonna get.