Relievers Are Not Worth Multi-Year Deals

While we’ve seen prices going up for free agents across the board, nowhere has the inflationary market been more clear than in the bullpen. After Joaquin Benoit set the market by landing a three year contrat with Detroit, teams have felt obligated to match the length of that deal in order to get the guys perceived to be the best bullpen arms on the market. Scott Downs, Matt Guerrier, and now Jesse Crain have all landed three year deals since Benoit signed, and Rafael Soriano is widely expected to get a deal of at least three years as well. After several years of restraint, teams have found themselves spending on the bullpen like its 2006.

Unfortunately for the teams with the open wallets, recent history suggests that giving contracts of 3+ years to a relief pitcher is generally a terrible idea. Here’s a list of free agent relievers who have received deals for three or more years since over the last four off-seasons.

Danys Baez (2007-2009), 3 years, $19 million: -0.4 WAR
Justin Speier (2007-2010), 4 years, $18 million: -0.2 WAR
Jamie Walker (2007-2009), 3 years, $12 million: -0.5 WAR
Scott Schoeneweis (2007-2009), 3 years, $11 million: -1.5 WAR
Chad Bradford (2007-2009), 3 years, $11 million: +2.0 WAR
Francisco Cordero (2008-2011), 4 years, $46 million: +2.8 WAR
Mariano Rivera (2008-2010), 3 years, $45 million: +7.8 WAR
Scott Linebrink (2008-2011), 4 years, $18 million: +0.5 WAR
David Riske (2008-2010), 3 years, $13 million: -0.6 WAR
Francisco Rodriguez (2009-2011), 3 years, $36 million: +1.7 WAR
Damaso Marte (2009-2011), 3 years, $12 million: -0.2 WAR
Brandon Lyon (2010-2012), 3 years, $15 million: +1.0 WAR

12 relievers were deemed worthy of long term deals as free agents. Half of them performed below replacement level and were a complete waste of money. Four of them were simply disappointments, ending up as semi-productive relievers making far too much money. One of them had a good first year, but still has to justify the decision with two years remaining on the deal. And the other is Mariano Rivera.

If you take Rivera out of the sample, as he’s just something totally different than the rest of these guys, the other 11 combined for +4.6 WAR over 30 player seasons. That’s an average of +0.15 WAR per year. It gets worse. K-Rod and Cordero, the other two closers in the sample, make up +4.5 of that +4.6 WAR. Yes, that means that the nine setup men who have signed deals for 3+ years since 2006 have combined for +0.1 WAR.

As a group, teams have paid for premium production and instead received the same level of performance that they could have expected if they had signed minor league free agents. The evidence couldn’t be any stronger: signing guys like Guerrier and Crain to three year deals is just throwing money away. It’s not that they’re bad pitchers; it’s that relief pitchers are so prone to huge swings in performance that trying to project the long term future of any of these guys is simply folly.

After going crazy in the winter of 2006, teams calmed down drastically on long term deals for relievers. But it appears that we’re right back in silly season mode, as more relievers have gotten three year deals this year than the last two winters combined. For all the talk about how major league teams are getting smarter – and they are – you’d never know it from the way money is being thrown around this winter.

Print This Post

Dave is the Managing Editor of FanGraphs.

81 Responses to “Relievers Are Not Worth Multi-Year Deals”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Pirateball says:

    The market is a joke right now for RP

    Vote -1 Vote +1

    • gnomez says:

      The market is a joke right now.

      Vote -1 Vote +1

    • MikeS says:

      Well, it’s a combination of supply and demand, expectation management, the evolution of the game, the way sabremetrics measures value and visibility.

      When a guy does his job and gets three outs in the 7th or 8th everybody just assumes that’s what he was supposed to do and he really gets little credit. It’s hard to rack up WAR pitching 2 or 3 innings a week and you don’t even get saves to show for it.

      On the other hand, give up two runs without getting three guys out just one time out of five and you stand out like a sore thumb. The fanbase wants to run you out of town. Maybe relievers need short memories, but fans and front offices never forget. Yet the average team still needs one or two of these guys to not spit the bit almost every night.

      Add to that the fact that just about all relievers are guys not good enough to be starters and set up men are guys not even good enough to be closers and you have a perfect storm of marginal pitchers who dissappoint all too often. If these guys could pitch 6 innings without giving up 3 or 4 runs they’d be in the rotation. Anybody who shows the glimmer of a hope of regularly getting three outs late in a game will be overvalued.

      Vote -1 Vote +1

  2. osfib says:

    To a certain a extent, I agree with your comments. However, would you have posted this article today, or at any point in the offseason, had the White Sox NOT sign any of the free agent relievers available to a multi-year contract?

    -51 Vote -1 Vote +1

    • Travis says:

      Your comment is nearly unintelligible. Are you suggesting that Dave published this article because he hates the Chisox? That’s absurd.

      +23 Vote -1 Vote +1

      • The Ancient Mariner says:

        Some folks on here, if Dave said the sky is blue, they’d insist it isn’t and proclaim it another example of bias (because after all, blue is one of the M’s colors, and so clearly he only said it because he’s a myopic Seattle homer).

        +15 Vote -1 Vote +1

      • neuter_your_dogma says:

        If you’re in Seattle, the sky is most likely gray :)

        +29 Vote -1 Vote +1

    • Bill@TPA says:

      That’s maybe the weirdest suggestion of bias I’ve ever seen. Now, would he have posted it had three of those crazy deals not happened within the past week? Probably not…typical FanGraphs topical-post bias

      +19 Vote -1 Vote +1

    • Reverse Vampires! says:

      MLB free agency analysis during the MLB free agency period!? I smell a conspiracy.

      +20 Vote -1 Vote +1

    • Kevin S. says:

      Because the White Sox not joining in on the parade of multi-year reliever deals would have made this article totally irrelevant!

      Vote -1 Vote +1

    • Small Sample Goodness says:

      I agree, Dave has a long, storied history of picking on the White Sox.

      What a poopy-head.

      Vote -1 Vote +1

    • Steve says:

      Dave, the only acceptable comments after a White Sox trade and or signing are as follows

      1. “Another, savvy, under-the-radar move by KW”
      2. “Cubs Suck”
      3. “Cubs Suck”

      +5 Vote -1 Vote +1

    • Steve says:

      While this post about the White Sox is obviously a little nutty, I will be interested to see how the “SABR” crowd reacts to Theo giving Bobby Jenks $12M/2.

      Vote -1 Vote +1

  3. Pat says:

    just this title made me giggle into my coffee.

    Vote -1 Vote +1

  4. Briks says:

    And in the meantime, the Rays are signing Peralta to a 1yr 900k contract. Its amazing how some teams just get it and some don’t.

    +5 Vote -1 Vote +1

    • chuckb says:

      exactly! And Bobby Jenks, who was recently non-tendered, is about to get a big deal as well. Imagine how much better the Reds would have been last year if they had spent those Cordero $ on a real left fielder rather than using Jonny Gomes day after day.

      Vote -1 Vote +1

    • jamie says:

      And they brought Howell back for a relative bargain. The Rays should be able assemble a perfectly good bullpen for about the same price as 1 year of Matt Guerrier. As should any team that tries hard enough, right?

      Vote -1 Vote +1

  5. Joe R says:

    A lot of Sox fans will be pissy about Theo not engaging in the RP market.

    But look at these prices. Terrible contract bonanza.

    Vote -1 Vote +1

    • Pat says:

      Agreed, until yesterday I was certain that Crain would end up in Boston. Guerrier’s questionable contract made it seem inevitable that Crain (who really is a better set up man) would land some thing similar. Although overly expensive, the Crain deal is marginally better than Guerrier’s.

      Vote -1 Vote +1

      • Shagomir says:

        I hope they enjoy the Crainwrecks. When he breaks down, Crain doesn’t have the ability to adjust, and will usually give up a whole mess of runs. Hopefully his new coach has a shorter leash than Gardy, I can think of 3-4 games just last year that Crain lost for the Twins when Gardy left him in there too long.

        The guy should be making minimum wage carrying baseballs to the umpire.

        Vote -1 Vote +1

    • Steve says:

      maybe WAR doesn’t “work” for relievers?

      not saying i have a better idea, but if the market keeps paying a certain price over and over for these guys, maybe we’re not looking at it correctly?

      i’m not criticizing the piece, just wondering if replacement level isn’t low enough for “non-closers”? the type of guys who are considered “freely available” middle relievers are minor leaguers who aren’t good enough to start or even close. in other words, they represent the bottom of the barrel in term of professional pitchers. maybe it’s not just so easy to grab a few of these guys and get the same result?

      i don’t know, just thinking out loud, feel free to tear me apart.

      +25 Vote -1 Vote +1

      • BEJ37 says:

        This is in line with my comment (below), which I sent before I saw this. I think guys like Downs, Benoit, Guerrier, and Crain are elite non-closing (set-up) relief pitchers — hard to come by, and not easily replaced.

        Vote -1 Vote +1

      • Erik says:

        I don’t much like Fangraphs WAR for pitchers, in general, because it’s exclusively based on FIP. For relievers, I think it’s even worse than it is for starters, but that doesn’t really change the point of Dave’s article. Just as a comparison, the number in parenthesis is each player’s WAR according to Baseball-Reference over the same period

        Danys Baez (2007-2009 -0.4 WAR (+0.1)
        Justin Speier (2007-2010) -0.2 WAR (+1.0)
        Jamie Walker (2007-2009), -0.5 WAR (+0.5)
        Scott Schoeneweis (2007-2009) -1.5 WAR(-0.6)
        Chad Bradford (2007-2009) +2.0 WAR (+3.2)
        Francisco Cordero (2008-2011) +2.8 WAR (+3.9)
        Mariano Rivera (2008-2010) +7.8 WAR (+10.2)
        Scott Linebrink (2008-2011) +0.5 WAR (+1.1)
        David Riske (2008-2010) -0.6 WAR (-0.5)
        Francisco Rodriguez (2009-2011) +1.7 WAR (+2.3)
        Damaso Marte (2009-2011) -0.2 WAR (+0.5)
        Brandon Lyon (2010-2012) +1.0 WAR (+2.0)

        Every reliever comes out better, but not enough to dispute the point of the article that relievers are greatly overpaid.

        +28 Vote -1 Vote +1

      • WY says:

        I agree 1000% with the above comments by Steve, BEJ37, and Erik about (Fangraphs) WAR and relievers.

        Vote -1 Vote +1

      • Jason B says:

        “hard to come by, and not easily replaced”

        Except they’re not, and they’re not. 0.15 WAR is almost the very defnition of replacement level – just 0.15 WAR away!

        Vote -1 Vote +1

      • Nathaniel Dawson says:

        Jason, I think you missed his point. He was trying to say that maybe those relievers really aren’t just +0.15 above replacement. He speculates that replacement level might be too high for these players, so maybe that average of +0.15 is really more like +1.0 or something. How low (or high) the bar is set changes the overall valuation.

        Vote -1 Vote +1

      • Cheese Whiz says:

        Or maybe if you just threw some of those minor league free agent or AAAA relievers into the bullpen you’d find that they gave you about the same result as the sample above. Maybe B-Rs WAR numbers are better than Fangraphs, but I’ve never seen a single compelling argument that these guys aren’t easily replaceable. Smart teams build bullpens for next to nothing all the time.

        Vote -1 Vote +1

      • Wally says:

        So a few things about some sort of new relievers replacement level:

        First, all pitchers are not still just pitchers, no? This isn’t really the same thing as awarding positional scarcity points for players that have the skill to field SS or catcher. Middle relievers are middle relievers because they kinda suck. It is self-selecting. Also, even closers, outside a few exceptions, are closers because they couldn’t be starters for one reason or another (they can’t throw 100 pitches at a time, they don’t have a quality 3rd or even 2nd pitch, they are only good against same-handed batters).

        I don’t see why we should award a lower bar to pitcher that come in during the middle innings just because they kinda suck. They still just pitch. They aren’t fielding a tough position where it is harder to find good batters.

        Second, relievers already have the advantage of not throwing to the same batters multiple times during a game. Its been clearly documented that relievers as a group have lower FIPs than starters probably largely for this reason, despite starters generally having more talent. Currently, WAR doesn’t adjust for this factor, right? So, they have a natural advantage there anyway, why give them a new artificial advantage?

        This idea that WAR for relievers might be broken looks good on its face, but I don’t really see any evidence that this is the case.

        Vote -1 Vote +1

      • CircleChange11 says:

        That’s what I was going to say.

        WAR doesn;t work for relievers, for a variety of reasons, including but not limited to limited innings pitched.

        This aspect is brought up every single time WAR and relievers are being discussed, yet relievers are continually being evaluated with WAR.

        It doesn;t make sense.

        Pick anything else, % of inherited runners scored, runs allowed (period), % of appearances allowing zero runs/hits/runners, etc.

        I don’t know that I necessarily have the best answer either, but I am confident that there must be something better than WAR.

        Talk about the anture of the position and how due to such small innings, stats can vary widely year to year, because everyone knows how important a good bullpen is. It is especially evident during a year when an otherwise good team has a poor bullpen. That’s Hell.

        Vote -1 Vote +1

      • Shagomir says:

        B-R uses a lower replacement level than Fangraphs, so their WAR totals are typically higher. It’s systematic, and it means nothing. An 8 win season using Fangraphs WAR is pretty much the same as a 10-win season using B-R WAR, from what I understand.

        The point is, don’t sign a reliever to a long-term deal.

        Vote -1 Vote +1

      • David K says:

        “Pick anything else, % of inherited runners scored, runs allowed (period), % of appearances allowing zero runs/hits/runners, etc.”
        My personal fave for relievers is RE24. It may not take into account ballpark effects or the quality of defense behind you, but i think it gives you a pretty good idea how much a pitcher can pitch his team out of, and into, trouble.

        Vote -1 Vote +1

  6. D4P says:


    Ned Colletti’s an idiot.

    Vote -1 Vote +1

  7. BEJ37 says:

    To play contrarian, let me ask, which team most over-achieved in 2010? If you said (as I might say) Padres, then you’d also probably agree that a key part of their surprising success was due to the strength of their bullpen, and not just their closer. To some degree, they may be an anomaly (taking advantage of Petco), but perhaps other teams view their surprising 2010 success as a harbinger of things to come, and/or a positive model to follow? Didn’t it seem like the Padres won an awful lot of close games last year? One non-closing relief pitcher, viewed in isolation, may not be (statistically) all that valuable but, in the case of a strong bullpen, maybe the whole is greater than the sum of its parts, and the seemingly free-spending GMs aren’t totally crazy?

    Vote -1 Vote +1

    • Briks says:

      But the Padres built that bullpen cheap. Heath Bell made 4 million in arbitration, and no other reliever on their team made over 1 million. And they all performed at the top end of their projections. But even if they hadn’t the Padres wouldn’t have had a lot of risk. I doubt if even the Padres thought their bullpen would be that good. But if teams really wanted to follow the Padres blueprint, they would be building their bullpens with young, cost controlled players, even future starters who are not ready to start yet (the way the Dodgers had previously done with Billingsley and Kershaw). You are likely to have an average bullpen at a below average price, with the chance that if everything breaks right you have a 2010 Padres situation.

      +8 Vote -1 Vote +1

      • Jack Str says:

        The problem with this seems obvious though, and you said it yourself: the Padres got lucky. Very lucky. In order for them to be as good as they were they needed most of their guys to overperform. That can’t be the smart way to build a good bullpen. That would be like building a team that’s going to play .500 ball most of the time and will occasionally make the postseason. That’s all right if you’re small market, but how should teams with more options proceed, and how should teams that are solid everywhere but their bullpens proceed?

        Vote -1 Vote +1

    • Erik says:

      The Padres excellent bullpen directly refutes the need to spend millions of dollars signing relievers to multiyear deals. The Padres assembled a fantastic bullpen incredibly cheaply, with only Heath Bell ($4 mil) and Mike Adams ($1 mil) making more than the league minimum. Teams SHOULD use the Padres as a model of how to build a bullpen…but that model says “The volatility and unpredictability of relievers allows you to build a good bullpen without spending much money or signing veteran relievers to long-term deals.”

      Vote -1 Vote +1

      • BEJ37 says:

        As I suggested in my comment to Steve, my impression is that elite set-up guys like Benoit, Downs, Guerrier, and Crain (and Adams and Gregerson) are nearly as rare, and possibly nearly as valuable, as closers. They are NOT dime a dozen, readily available through most teams’ farm systems. Maybe the Padres were just lucky, or maybe Towers et al. are exceptional at spotting talent (interesting to see how the D’backs pen does, next year), but I’m not convinced that other teams can match (or even come close to) the Padres’ success on the cheap.

        Vote -1 Vote +1

      • Jason B says:

        Erik just totally nailed it. Totally. Nailed. It. Middle-relievers are so, so very fungible. Use them while their cheap and then trade them away to teams who will overvalue/overpay for that skill set, or release them when they get expensive and start all over with another bunch who can accomplish the same at a fraction of the price.

        Vote -1 Vote +1

      • chisxo24 says:

        I think teams are also looking at Arizona’s bullpen performance and seeing the flip side of the coin. And obviously, this is part of volatility. Teams seem to be approaching free agent reliever evaluation with the idea that pitchers with a proven track-record (or even just one good year) are inherently less risky. Teams want to smooth the volatility of performance, and this is the only answer that they see. Obviously, you can debate whether free agent relievers are less volatile at all, especially over a period of 3 years.

        Vote -1 Vote +1

      • Cheese Whiz says:

        The other issue is simply sample size. Most relievers don’t get more than 50 innings a year. In a sample size that small almost any result is possible unless the player is a truly elite talent, which middle relievers by definition are not. There is simply no way to guarantee a good return on your investment with that kind of volatility.

        Vote -1 Vote +1

  8. striker says:

    Relievers are hard to gauge because ERA generally doesn’t work in their favor. One bad outing (1IP 4ER) and 3 good outings (3IP 0ER) can still result in a 9.00 ERA.

    In 2010 Crain pitched in 71 games. Here is the breakdown of games by ER.

    0ER – 60
    1ER – 4
    2ER – 3
    3ER – 3
    4ER – 1

    So 60/71 or 85% of the time he takes the mound he doesn’t surrender a run.

    64/71 or 90% of the time he takes the mound he surrenders 1 ER or less.

    From an innings perspective:
    0ER Outing – 58IP
    1ER Outing – 3.1IP
    2ER Outing – 3.2IP
    3ER Outing – 2IP
    4ER Outing – 1IP

    So 58/68IP or 85% of the innings pitched he doesn’t give up a run in that outing.

    It’s like relief pitchers need the equivalent of the Quality Start. Scoreless Outing, and Scoreless Outing Percentage.

    Or better yet, to gauge all pitchers come up with the % of innings they pitch and don’t get scored upon. The higher the percentage the better the pitcher.

    +5 Vote -1 Vote +1

    • Erik says:

      That would fail to account for inherited runners, which relievers should bear some of the responsibility for (whereas, with ERA, they have no responsibility for inherited runners).

      Vote -1 Vote +1

    • ToddM says:

      And yet, that’s often not even the job of the reliever in question. I followed Crain pretty closely last year, so this isn’t a knock on him, but relievers that DON’T always start innings definitely can’t be judged this way.

      If the guy comes in with 1 out and runners on first and second and gives up a slow grounder to advance the runners and then a single that plates both of them, he screwed up. Unless that last hitter manages to score from first with two outs (not likely), the reliever’s ERA is zero, but the guy he replaced gets tagged. The same things happen in reverse to relievers all the time.

      You could check out FanGraphs WPA (Win Probability Added) for relief pitchers to account for a lot of that, but even then, different relievers pitch in difference leverage situations, so even those numbers are a bit misleading at times.

      I do agree on one thing — FanGraphs’ WAR, especially for pitchers, and even more especially for relief pitchers, undervalues the top setup men and closers. Even so, a lot of these guys are getting seriously overpaid. 3 and 13 for Jesse Freakin’ Crain is unbelievable.

      +11 Vote -1 Vote +1

  9. U-G says:

    3 years might have been a bit much, but $13MM over it for Crain really doesn’t seem that bad.

    Vote -1 Vote +1

    • Ricky says:

      Top notch analysis.

      -7 Vote -1 Vote +1

    • Cheese Whiz says:

      It is when you could get the same performance for league minimum.

      Vote -1 Vote +1

      • CircleChange11 says:

        Go ahead and tell me, right now, which 5 relievers are going to deliver top notch relief performances in 2011 for league minimum.

        We know there will be some, I’m asking you to name them, so that GM’s know which guys to sign.

        Vote -1 Vote +1

  10. Felonius_Monk says:

    In general I agree with the point of this article (relievers, in most cases, are over-priced), but I’m not sure basing it on the (IMO extremely flawed) FG FIP WAR figure (which, if I’m correct, uses the, again IMO, ostensibly ill-thought-out “chaining” philosophy to correct for leverage) is the best way to demonstrate that.

    Vote -1 Vote +1

    • WY says:

      I agree. Some Fangraphs writers like to trumpet the correlation between Fangraphs WAR and team performance, with Fangraphs win values and free-agent signings, and so on. Yet their model is clearly out of line with the reality in terms of how teams actually value relievers. Rather than examining their own model, the response of (at least some) of these Fangraphs writers is to conclude that the GMs who make these deals–who know much more about baseball than anyone who writes for Fangraphs–are idiots.

      Vote -1 Vote +1

      • WY says:

        “That last sentence should read, “Rather than examining their own model, the response of (at least some) of these Fangraphs writers is to conclude that the GMs who make these deals –- who know much more about baseball than anyone who writes for Fangraphs –- are idiots.”

        There was some reformatting done with a couple of the em-dashes…

        Vote -1 Vote +1

  11. MFG says:

    Is Jason Frasor gonna end up closing in Toronto?

    Vote -1 Vote +1

  12. I just looked at 5 of these guys (Speier, Baez, Riske, Linebrink) and none had the talent to begin with. They all had career xFIP before the contract above 4.00 and got paid for small sample sizes where thye looked to dominate.

    This is a case of teams looking at stats like ERA and paying based on that. I would like to see WAR for these players the three years leading up and see how much difference they were from projections for the next three.

    Vote -1 Vote +1

  13. Tom B. says:

    It seems that the sample is only of free agent signings and excludes contract extensions. While it may be legitimate to separate out these groups (contract extensions may skew more towards younger relief pitchers entering peak seasons, no one’s in the market for someone who’s already signed to another team), it may help paint a fuller picture to examine RP’s who’ve gotten a 3+ year contract extension before saying RP’s don’t deserve multi-year deals.

    The contract extension that immediately leaps out to me is Joakim Soria, who signed a multi-year extension early in the 2008 season and has gone on to post 5.5 WAR in the past three years (although a portion of the 2008 number came before he signed the extension)..

    Vote -1 Vote +1

  14. Jacob says:

    I’m fine with the opinion that not all relievers are worth multi-year deals, but using WAR to prove the point doesn’t really do much for me. It’s a heavily flawed statistic and there is too much room for interpretation when calculating WAR.

    There are more things to take into account when signing relievers. In some cases, it may be more cost effective to solve bullpen issues internally, but there is higher risk involved with using a pitcher from your farm system who has no MLB experience than to sign someone with an MLB track record. Free agency is an earned privileged and just the fact that the pitcher is on the open market to begin with shows consistency at the major league level.

    No team is forced to sign a reliever to a multi-year deal, yet it happens when the alternatives are limited.

    Vote -1 Vote +1

  15. Blue says:

    It’s all about the free supplemental picks, baby.

    Vote -1 Vote +1

  16. David says:

    I wonder if we’re underestimating the cost of wins on the free market in general. It makes no sense to give a reliever a three year contract. It also makes no sense to give Johan Santana or CC Sabathia or Jason Werth seven year contracts.

    The open market gives more years and money than I’d rationally expect across the board. Not just to relief pitchers.

    Vote -1 Vote +1

  17. Dan says:

    The Twins are going to be a perfect case study. We will see how well their bullpen performs next year (and subsequent years) without Crain, Guerrier, Rauch, Fuentes, et al. It will be interesting to see who they have pick up those innings.

    Vote -1 Vote +1

  18. Blue says:

    Despite my snark above, here’s what I think is going on.

    Most teams are pretty set in position players and starting pitchers. When they perceive themselves to be close to contention and a handful of wins could make a difference replacing two or three replacement level arms in the bullpen is just a heck of a lot easier than trying to upgrade somewhere else. There’s ALWAYS room in a bullpen for another good arm while adding a 1 or 2 win upgrade at a position is a bigger pain in the butt.

    Vote -1 Vote +1

  19. Is the Jenks signing going to be among the worst signings of this offseason? At least it’s only 2 years, but wow, $6M, poor Daniel Bard!

    Vote -1 Vote +1

    • williams .482 says:

      why “poor daniel bard”? this just means that the Red Sox have a Proven Closer(TM) to step in when Papelbon leaves as a free agent, leaving Bard in his current and much more valuable fireman role.

      Vote -1 Vote +1

  20. PG says:

    This stuff about relievers being overpaid is almost as annoying as the complaints about it NOT being the year of the pitcher. We get it.

    But anyways, the White Sox needed a reliable RHR, and the free agent market for relievers has exploded. They have Sergio Santos and Tony Pena. Santos stuggled in the second half and has been pitching for a yeah and a half. Pena just isn’t really that good.

    I’m sure Ken Williams knew he was overpaying, but after declaring he’s going “all-in” and handing out big contracts to Dunn and Konerko and resigning Pierzynski, would it really make sense for him to rely on some retread? Or wait for the market to correct itself and risk ending up with nothing? Not really.

    Vote -1 Vote +1

    • MorneauVP says:

      Who’s to say Crain is anything more than a re-tread? He had four good months, but he was nothing special from April 2007 – June 2010.

      Your perception of Crain is the precise element that’s wrong in the analysis of relievers that leads to these deals. You say Sergio Santos struggled in the second half, but that sample size is barely smaller than the one you use to insinuate that Crain is a slam dunk.

      Vote -1 Vote +1

      • PG says:

        I never said Crain was a slam dunk. I’m saying he’s more reliable than someone like Manny Delcarmen, for example.

        And Santos’ poor second half can’t be treated the same as a poor second half by the average relief pitcher. Santos didn’t start pitching AT ALL until the tail end of the 2009 season, and the results in the minors weren’t pretty. He had an incredible start to the season, but it was obvious that the league caught up to him as the season wore on. To count on him as your primary RHR would be foolish.

        Vote -1 Vote +1

  21. MorneauVP says:

    I’m shocked that teams haven’t wised up to this concept yet, especially in light of these deals and the success a league-minimum bullpen like San Diego had in 2010. The Rays seem to be doing a good job, but we’ve seen teams spend even more freely on marginal talent than usual.

    Jesse Crain had a tremendous 4-month run. I watched every outing first-hand, as a Twins fan, and yes, he was dominant. But this is a guy with a tremendous history of injury, and the White Sox (like the Dodgers, Tigers, etc.) have completely missed the point of sample size.

    You see it often times with position players or starting pitchers (recent examples include Adrian Beltre, Carl Pavano), where a guy who’s had some bad luck will sign a one-year deal with incentives and try to position himself for a better contract next season. Sometimes it works, sometimes it doesn’t (Sheets, Harden). But even the successful players come with marked risk — their success is very recent, over the course of one season, and you’re looking to commit for three, four, maybe five seasons. Will a team like the Angels pay Adrian Beltre for five years based on one good season? Probably. Should they? I think conventional wisdom says no.

    And yet with relievers, the sample size on which they’re based is even more microscopic, thus escalating the risk exponentially. If Chien Ming Wang had come back at the end of last season to make eight or nine strong starts and totaled 55 innings, would he have suddenly netted a multiyear deal? Doubtful. His health and success would be questioned because it was limited to only nine appearances. Yet this concept is still lost on the relief market.

    On June 5, Crain’s ERA sat at 5.47. He was in the midst of a miserable season. His next 42.1 innings saw him allow four earned runs and strike out 42 hitters. He held opponents to a ridiculous .163/.272/.191 line. He was utterly dominant. And those 42 innings earned him $13M.

    In what other realm of baseball are players judged on such an absurdly small sample size? Could Jermaine Dye have signed on with the Pirates and raked for the final five games of the regular season (45 innings), and then gone on to sign for three years? Jesse Crain threw 1099 pitches last season. Joaquin Benoit threw 941. Matt Guerrier threw 1045.

    That’s about as many pitches as a starter will throw over the course of six weeks, and yet GMs treat such small numbers (Hell, even smaller in Crain’s case, since his first two months were awful) as the basis for handing out $10M-$16M.

    There’s simply no logic in it, and it shows through the volatility of their performances. Relief pitchers will always be unpredictable, because their sample size over the course of a season is nothing more than a slump or a hot streak in a starting pitcher’s standard season. Shelling out money and expecting them to be consistent, then scratching your head when that’s not the end result is lunacy. Kudos to the GMs who shy away from it. Your money is better spent elsewhere.

    Vote -1 Vote +1

    • CircleChange11 says:

      Great post. Sample size, IMO, is the key issue. It’s the equiavelnt of basing a batter’s season on one month’s worth of data.

      We say crazy things like relievers are fungible or volatile or can be siubsituted for replacement level guys, etc.

      What we really need is a “Quality Appearance” stat or something of that nature which looks at how many appearances a reliever made, and quantifies how many of those were situations where they “did a good job”.

      Due to small IP numbers, a soingle outing can have to great of an effect on the overall.

      We know this for every other player, except I guess, for relievers. Seems like we know better, but choose not to follow it through.

      At this point, I wonder if it is because it feeds our bias that relievers are just failed starters (I hate that expression) that can replaced with anyone “off the street”.

      Vote -1 Vote +1

      • Wally says:


        This “Quality Appearance” stat, would still be subject to the same sample size issues, and not resolve the general criticisms of evaluating relievers, if we assume those criticisms are valid in the first place.

        Vote -1 Vote +1

      • CircleChange11 says:

        In the same regard that “everything” is subject to small sample size for relievers.

        The point is, with something like ‘Quality Appearance’ is that one horrible outing doesn’t ruin or skew the season stats.

        A reliever could pitch 6 scoreless innings over 6 games, and then give up a 3-run homer in the 7th, and most metrics would represent the season stats as “average”. Now, how the 6 out of 7 (quality appearance) stat compares depends on what the stats are of other relievers. If the average reliever has a quality appearance 2 out of 3 times, then we have our baseline comparison.

        It’s really not that difficult. I don’t think we have a good method of evaluating relievers yet.

        I think the first that has to happen is for us to stop viewing relievers as the second-class citizens of the pitcher community. All relievers do not fall under the same umbrella. There’s quite a bit of difference between the long reliever whose 5.42 K/9 stats fall under the same heading as the flamethrowing reliever whose 10.54 K/9 renders him dominant.

        When it comes to sabermetrics, player evaluation, and relievers, I don’t think we’re very accurate at all. I think we often get the conslusion we want in order to hold onto beliefs and opinions that we enjoy having … especially the comments/research about using the ace reliever in the 7th if that’s where the highest leverage situation is (especially when we can look back with hindsight and easily see when the highest LI situation is). It’s not see easy to see in the 7th inning that NOW is the highest LI of the game. Duh.

        Vote -1 Vote +1

  22. CarlosM7 says:

    If I told you how easy it is to get a job in this recession, you wouldn’t believe me. But the truth is more employers are going online to find people just like you and me who are ready to work at a good job (one that pays good!). The only thing that makes sense is to stop wasting time driving around all day filling out a dozen applications and going from one boring low paying job to another. I found this site that pretty much matches you up with your dream job that is available in your city right now. I have found it very helpful. Go to

    -7 Vote -1 Vote +1

  23. WAR works for relievers the same way FIP and xFIP do. The guys at FanGraphs know this.

    Vote -1 Vote +1

  24. Doesn’t anyone else find it interesting that the author picked “three or more” years instead of any “multi year” contract, as per the article title. I guess this was to avoid including the multi year success stories of Jose Valverde, Kerry Wood, Brian Fuentes, etc.

    Fact is, most multi-year (multi being “more than 1”)contracts are not “worth it”, this does not hold specific for relievers, but all MLB free agents – see Raul Ibanez, Milton Bradley, Pat Burrell, Oliver Perez.

    Vote -1 Vote +1

  25. Koch says:

    Am I the only one who can’t stand CircleChange11.

    Vote -1 Vote +1

  26. Lanidrac says:

    How do 2 year contracts not count as multi-year contracts? You need to change the title of this post.

    Vote -1 Vote +1

  27. Bang Bang says:

    CircleChange11 is a fucking cocksucker.

    Vote -1 Vote +1

  28. Kyle says:

    Teams panic if they have a bullpen that isn’t clutch with a guy for the 7th inning, a setup man, and a closer. I think this is what we’re seeing… the Red Sox made how many offers to relief pitchers for multi-year deals? After they had made a huge splash with free agents(but when you have Lackey, Dice-K, and the wild-card Beckett), maybe it’s pretty smart. The Yankees then followed suit…

    But teams who’ll give Burnett, Lackey, and others crazy contracts with their best years behind them with money to burn… but that Cordero contract looks insane. Not to mention why’d the Red Sox give Jenks so much money when they already have Papz in decline apparently. Will Jenks make Papelbon actually fear for his closer job and bounce back? Doubtful… but I can see why they offered Rivera so much. He won’t drop off until he’s about 45 years old?

    Vote -1 Vote +1

  29. Erin Andrews says:

    I intended to post you the tiny remark to help give many thanks once again regarding the unique advice you have shared above. It is so seriously open-handed of you to allow freely just what most people could have marketed for an electronic book in making some cash for their own end, certainly given that you could possibly have tried it if you ever desired. Those techniques also worked to become a good way to realize that other individuals have the identical passion just like mine to know a whole lot more with reference to this problem. I’m certain there are lots of more pleasurable instances in the future for many who looked at your site.

    Vote -1 Vote +1

  30. Ryan says:

    I clicked through to this article after reading the story on Paplebon signing with the Phillies. This article speaks for itself, I would rather take my chances on three guys in the $2-$3M range, as opposed to going “all-in” on one “closer” who makes $10M+.

    As a Phillies fan, I am compelled to ask, why is Brad Lidge not on your list of bad contracts. I believe his WAR from 2009-2011 was right around zero, yet he was paid almost $40M for those years on the deal he was signed to in mid-2008.

    Vote -1 Vote +1