Multi-Year Deals for Relievers: An Ugly Retrospective

Today, the Reds placed Jonathan Broxton on the disabled list with an elbow injury that will require season ending surgery, bringing to an end a disappointing year that began with a pretty nifty paycheck over the off-season. Broxton is hardly the only relief pitcher who got paid last winter and hasn’t quite earned his salary this season, however. Below is a table noting the performances of the 13 relievers who signed mutli-year contracts over the winter.


Name Years AAV Total IP ERA FIP xFIP WAR RA9-WAR
Jason Grilli 2 $3,375,000 $6,750,000 42.1 2.34 1.62 2.07 1.6 1.2
Randy Choate 3 $2,500,000 $7,500,000 28.0 2.25 2.46 3.30 0.5 0.8
Tom Gorzelanny 2 $2,850,000 $5,700,000 36.2 2.45 4.26 3.97 -0.1 0.7
Joel Peralta 2 $3,000,000 $6,000,000 53.0 3.23 3.54 4.09 0.5 0.6
Sean Burnett 2 $4,000,000 $8,000,000 9.2 0.93 4.17 3.54 -0.1 0.6
Rafael Soriano 2 $14,000,000 $28,000,000 53.1 3.71 3.93 4.14 0.2 0.3
Jeremy Affeldt 3 $6,000,000 $18,000,000 33.1 3.51 4.44 4.49 -0.5 0.2
Joakim Soria 2 $4,000,000 $8,000,000 13.0 3.46 3.42 4.59 0.2 0.2
Mike Adams 2 $6,000,000 $12,000,000 25.0 3.96 5.23 3.79 -0.4 0.1
Kyuji Fujikawa 2 $4,500,000 $9,000,000 12.0 5.25 2.78 2.85 0.2 -0.1
Jonathan Broxton 3 $7,000,000 $21,000,000 30.2 4.11 4.66 4.50 -0.4 -0.3
Brandon League 3 $7,500,000 $22,500,000 44.2 5.24 5.18 4.23 -1.0 -1.4

You may note, perhaps, that the low AAV guys are at the top — the good half, since the list is sorted by RA9-WAR — and the high AAV guys are at the bottom. The two worst performers on the list are two of the four guys who got three year deals, as Broxton and Brandon League have been nothing short of unmitigated disasters. The guy with the highest salary, by far, has also been pretty disappointing, and Rafael Soriano is one of the reasons why the Nationals haven’t been particularly good this season.

Basically, of the 13 multi-year deals for free agent relievers, you could point to four of them that seem like good ideas in retrospect: Grilli, Choate, Peralta, and Soria. You could also maybe throw Gorzelanny into that mix if you wanted, especially given his service in the rotation. All of them were low salary guys, getting a second year (or a third in Choate’s case) at a rate that wasn’t so far above the league minimum that the risk was still minimal, and each one has performed pretty well overall, with Grilli being one of the steals of the winter.

The high cost guys, though, have been pretty much atrocious across the board. League has been perhaps the worst relief pitcher in baseball. Broxton was bad and is now hurt. Adams was bad and is now hurt. Affeldt has been okay, but nothing special. Soriano has been mediocre and wildly overpaid, plus he cost the Nationals a draft pick. Burnett and Fujikawa both got hurt and have hardly pitched. The failures clearly outweigh the success stories.

All told, these 13 pitchers have combined to throw 382 innings this year, putting up +0.8 WAR/+2.9 RA9-WAR. Even if we take the more favorable runs allowed number, that’s a total of +3 WAR for a combined 2013 cost of $65 million, or about $22 million per win! That’s, uhh, not a very good price.

None of this is news, necessarily, but the numbers paint a stark picture of the risks of spending real money on free agent relief pitchers. They’re the worst deal in baseball. Their performance is unpredictable, and even the guys we think have established themselves with some kind of track record are capable of just turning into total lemons.

It is possible to sign a smart multi-year contract for a relief pitcher. It just probably shouldn’t be for a significant amount of cash.




Print This Post



Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.


74 Responses to “Multi-Year Deals for Relievers: An Ugly Retrospective”

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

    not sure, but the WAR for Gorzelanny is incorrect, as are the IP. When were these stats taken from?

    Vote -1 Vote +1

    • IZZY2112 says:

      He made 9 starts. It may just be the statistics from his relief appearances.

      Vote -1 Vote +1

      • robby says:

        fair enough, though that shouldn’t take away from his overall value. He was signed as a swingman, so it’s not like he isn’t earning his pay or valuable when he’s not relieving…

        +7 Vote -1 Vote +1

        • Iron says:

          So including starts his WAR would go down from 0.7 to 0.4. I think it is less of not ‘giving him credit’ as it is not penalizing him.

          Vote -1 Vote +1

        • Craig says:

          Iron, his WAR goes from -0.1 to 0.4 including his starts. He has -0.1 in relief and 0.5 as a starter.

          Vote -1 Vote +1

        • SKob says:

          “You could also maybe throw Gorzelanny into that mix if you wanted, especially given his service in the rotation.”

          The author is giving him credit and would also lean towards throwing him into the group of good values. Starting is separate. Let it go!

          Vote -1 Vote +1

  2. Baltar says:

    This is just astounding, Dave. I thought the Rays’ signing of Peralta was one of the all-time steals, but he has barely been worth his pay at best so far–and that was one of the best deals!
    The Nationals must have forgotten to take their meds the day they signed Soriano.

    Vote -1 Vote +1

  3. Wil says:

    I think of Kimbrel when I think of the volatility of relievers. Will the Braves take a chance, and if they do will he be worth the money?

    Vote -1 Vote +1

    • Iron says:

      Yes and no.

      Vote -1 Vote +1

    • Larry Yocum says:

      Kimbrel may be a different animal though. He isn’t your average RP. As long as he stays healthy (isn’t this always the case for pitchers*), he should continue to be a monster out of the pen. Of all the guys currently closing, Kimbrel along with Chapman may be the ones primed to take the mantel of best closer after Mariano retires this year. All of the guys on this list were pretty average middle men when they were signed and Soriano was grossly overpaid for a guy with limited closing experience and a history of arm issues.

      Vote -1 Vote +1

      • TKDC says:

        Call me a Braves homer if you like, but I think the guy who leads* all relievers since 2010 in ERA, ERA-, FIP, FIP-, xFIP, xFIP-, tERA, SIERA, K% and WAR is already the best reliever in baseball, no offense to the greatest reliever in the history of the game.

        *many of these by huge margins

        Vote -1 Vote +1

        • TKDC says:

          And for the traditionalist among us, also saves!

          Vote -1 Vote +1

        • Bip says:

          He has a long way to go before he’s the best reliever of all time, but there’s no way to argue that Kimbrel isn’t the best reliever in the game today.

          Vote -1 Vote +1

        • Larry Yocum says:

          Well, I was kind of implying that he would be well worth the money by comparing him to Rivera in the first place, so you won’t get an argument from me.

          Vote -1 Vote +1

        • SKob says:

          @Larry

          Or you could have been implying the Yankees would just buy him when he was a free agent!

          Vote -1 Vote +1

    • Tim says:

      I hope the Braves dont spend 15 million a year on Kimbrel and I don’t think they will. Frank Wren is great at putting together cheap, good, bullpens so I’d rather they spend that money to lock down Heyward/Simmons/Freeman or possibly McCann

      Vote -1 Vote +1

      • Brendan says:

        Agreed. Even if Kimbrel performs exactly as he had previously a medium payroll team can’t pay 15-20% of their payroll on a reliever who pitches 65 of 1500 innings.

        Vote -1 Vote +1

        • Moves Like Munenori says:

          They won’t. They’re like the Cardinals right now, they have tons of young arms without enough places for all of them. The only reason Alex Wood and Julio Teheran even got a chance this year was because of injuries…if Kimbrel demands too much, Venters or someone else will take his place.

          Vote -1 Vote +1

        • Snowman says:

          Someone would take his place, but I’d pretty much bet against it being Venters at this point, at least for any significant amount of time. The Bravos Scott Proctored him with those 166 appearances in 2010 and 2011.

          Vote -1 Vote +1

      • Dan Ugglas Forearm says:

        They won’t spend too much on Kimbrel. I think even if they start to sense that his arb numbers are going to get ridiculous, they’ll trade him to a team in need of a Proven Closer. What separates the Braves bullpen from the rest is their lack of reluctance to turn a SP prospect into a bullpen arm (shout out to the Cards for doing similar things). If Kimbrel gets too pricey, welcome to the Bigs, Jason Hursh. They stockpile arms that are almost sure things in the bullpen, and take whatever they get from the ones who end up being able to start.

        Vote -1 Vote +1

    • Anon21 says:

      I’ve been hoping they’ll take a chance… by shipping him out for some young, cost-controlled talent. The free-agent prices teams are willing to pay for “proven closers” suggests they ought to be able to pull off a pretty good deal… although perhaps those “proven closers'” underperformance has already popped the bubble.

      Vote -1 Vote +1

      • Oh, Beepy says:

        Something tells me that most FOs would not value Craig Kimbrel anywhere on the generally accepted spectrum of reliever value. He’s sort of like Stanton in that regard, he’s a prodigious talent who is polished beyond his young years and is considerably more of a ‘sure thing’ than most other players with his particular skillset.

        Vote -1 Vote +1

        • Anon21 says:

          True. I am sure there are many FOs who would wisely stay away, but it only takes one fool betting on the myth of the proven closer to translate into a very good haul for a guy who only pitches 60 or 70 innings per season. I really hope the Braves explore a Kimbrel trade this winter.

          Vote -1 Vote +1

        • Dan Ugglas Forearm says:

          I don’t doubt the Tigers would do Castellanos/Rondon, and maybe even Porcello.

          Vote -1 Vote +1

  4. eddiegaedel says:

    I am recalling back when Soriano signed with the Nationals, pundits at ESPN said it was all that the Nationals needed and was a move for the postseason rather than in-season. Looks like Clippard could have handled that closer job and the team results wouldn’t be that different. Not sure where they would have re-allocated the money but it is seeming like a misplay on the Nationals part.

    Vote -1 Vote +1

    • Moves Like Munenori says:

      ESPN analysis is always spot-on. They would have crushed the Nationals if they had let Soriano go too, and in some way lumped it with limiting Strasburg’s innings to say that the front office didn’t want to win. I loathe ESPN.

      Vote -1 Vote +1

    • Dan Ugglas Forearm says:

      I’m not sure the Nats could have fielded a much better team. You can’t necessarily plan for injuries or underachievements of that magnitude (Espinosa, Span, Zimmerman on D). The only position I think they would have liked to upgrade would have been C. But there aren’t exactly lines of starting caliber catchers waiting for offers in the offseason.

      Vote -1 Vote +1

  5. triple_r says:

    With regards to the Nats and Soriano, they looked like geniuses at first, when they let two of the guys on this list (Burnett and Gorzelanny) walk, only to do a vexing about-face and sign the most expensive guy of the bunch. Regardless, he’s one of the main reasons for their disappointing season.

    Vote -1 Vote +1

    • Euruproktos says:

      You’re right. He is tied for worst offensive production on the team.

      Vote -1 Vote +1

    • Moves Like Munenori says:

      You forgot Michael Morse…they were ridiculed for letting him go too.

      Vote -1 Vote +1

    • Mcneildon says:

      How could a relief pitcher be one of the main reasons for a sub-.500 team’s disappointing performance?

      Vote -1 Vote +1

      • Snowman says:

        Soriano has 6 blown saves and 3 losses. Even were one to assume that none of the losses coincided with the blown saves AND that whoever took his place would have been Gagne-perfect all year (both ridiculous assumptions, obviously), those 9 extra wins would still leave them 5 games out in the division and a half-game out of the second wild card.

        Vote -1 Vote +1

  6. Stolmy Pimentel says:

    The Padres contract with Huston Street, which Dave criticized in itself and in relation to the Padres’ unwillingness to pay Adrian Gonzalez, is a similar case of good money gone bad, or to not sufficient good. I liked the Pirates’ acquisition of Melancon+ for Hanrahan before the year and argued my brother, also a Pirates fan, about it. Although the advantage has been aided by an unlucky injury, it too is a case for choosing low cost relievers with upside.

    Vote -1 Vote +1

    • Ivan Grushenko says:

      Swapping Gonzalez for Rizzo (and other stuff) is defensible, if the cost savings is used on something other than Huston Street.

      Vote -1 Vote +1

  7. Mac says:

    The biggest shock for me has been the success of Grant Balfour.

    The A’s long ago figured out reliever volatility (there’s a section of Moneyball explaining how teams overpay for RP help). Oakland knows to avoid these deals. But a funny thing happened in the 2010-2011 offseason. The free agent market completely dried up when all of Beane’s targets signed with other teams. January comes with $10 mil burning a hole in Beane’s pocket. Beane decided that shoring up the bullpen was better than spending no money at all and picked up Brian Fuentes ($5 mil AAV) and Grant Balfour ($4 mil AAV) on 2 year deals.

    Fuentes goes down as one of the worst A’s signings ever. The main legacy of that deal was that Bob Geren’s misuse of Fuentes and the subsequent atrocious pitching got the manager sacked, which led to the far superior Bob Melvin taking the helm.

    Meanwhile Balfour has so far compiled 2.8 WAR over his 2011-2013 tenure with the A’s. That deal has stunningly paid off.

    Big money for relievers is a high risk, low reward game but sometimes you do beat the odds.

    Vote -1 Vote +1

    • Justin says:

      And if you go by RA9 WAR Balfour looks even better, putting up 6.1 WAR in that same time frame.

      Vote -1 Vote +1

    • Guy says:

      I thought Balfour was a great signing at the time, and that Fuentes was a horrible signing that would end up just the way it did. Guess the A’s should sign me to scout their relief FA signings. Not only that, it’s not a bad commute for me.

      Vote -1 Vote +1

  8. SoxFan says:

    Yeah you forgot Koji who became multi-year with vesting option though

    Vote -1 Vote +1

    • Michael says:

      And Breslow wasn’t a free agent, but received a multi-year deal at comparable cost and has pitched well.

      Vote -1 Vote +1

    • Henry says:

      right but the point of this was to show that signing people to multi-year deals without performance incentives is dangerous, so yeah, uehara and breslow have both been good, but not totally relevant to this conversation.

      Vote -1 Vote +1

  9. Shauncore says:

    “League has been perhaps the worst relief pitcher in baseball.”

    Woh woh woh woh there. Carlos Marmol takes offense to that.

    +8 Vote -1 Vote +1

  10. Cards Fan says:

    Don’t forget about Jason Motte and his 2 year, $12M deal ($6M AAV). Only took a few months to require Tommy John surgery.

    The best part was that it was an arbitration deal and didn’t require multiple years.

    Vote -1 Vote +1

  11. Robbie G. says:

    Why do teams keep giving big money deals to relievers? How much more data do you need to reach the obvious conclusion that it’s a terrible idea? Bizarre!

    Vote -1 Vote +1

    • NBarnes says:

      I think part of it is emotional. Losing a game because your bargain bin closer gave up two runs in the 9th hurts more than losing a game because your #4 starter gave up two runs in the 5th. The latter is just the sort of thing that happens. The former gets headlines in the Sports page of the local newspaper and uncomfortable questions at the postgame press conference.

      Vote -1 Vote +1

    • AC of DC says:

      I would guess that it’s also at least in part a market matter. Somebody is willing to pay the Free Agent reliever too much money, and in all likelihood that’s going to decide where the guy signs. If you are a GM who refuses to overspend for relief, then unless you can pull off a number of savvy trades for good relievers under cheap contracts, your bullpen is going to consist largely of guys brought up from the minors, meaning probability suggests you’re going to be fielding a lot of replacement level players. Better, some would contend, to gamble on an established reliever repeating performance than on a kid from the farm turning out to be superior.

      Vote -1 Vote +1

  12. Stan says:

    Isn’t this just a function of relievers having low WARs in general? Couple that with a limited subset and the results are obviously not going to be pretty. Last year, there were only 13 relievers who had >1.5 WAR. Any GM that gives a reliever a >7M AAV contract is essentially hoping that guy is a top 10 reliever.

    Ergo, Ruben Amaro is crazy.

    Vote -1 Vote +1

    • Bip says:

      WAR will always underrate the best relievers, because they provide their value in a context where preventing runs is most valuable. Most other types of players will have an average leverage over their playing time over the course of a season. That is not true with good relievers only (or dedicated pinch hitters, if there was such a thing.)

      Vote -1 Vote +1

  13. Bip says:

    You shouldn’t be looking at WAR for this. WAR is a good way to evaluate player skill, because it accounts for what a player can control, factors out what he can’t, and it’s context independent. However, we aren’t looking at player skill, we’re evaluating player value. Contracts don’t pay players for skill, it pays for value contributed towards winning.

    For starters and position players it makes sense to conflate player skill and player value. Their contributions come independent of context; they play in close games and they play in blowouts. Context does not affect when they play. This is not true for relievers, however. The more effective relievers are payed to come into games when the leverage is highest. A scoreless inning pitched by a closer is more valuable than a scoreless inning by a starter, almost always. Since their performance is magnified, the value of being a sub-2 ERA reliever is magnified as well, compared to being an average reliever in the same context.

    Because effective relievers play in high-leverage situations exclusively, their value added is greater than WAR specifies. A context dependent measure like WAR is not appropriate for evaluating how well a contract turned out. On the other hand, we would use something like WAR to evaluate what sort of contract to give a reliever, because a team wants to know that players skill; the team will provide the context.

    Vote -1 Vote +1

    • Bip says:

      So we would want to use WPA, in other words.

      Vote -1 Vote +1

      • Darren says:

        Well WPA with a Replacement Level adjustment. WPA is above or below average.

        Vote -1 Vote +1

      • Michael says:

        I’ve heard WPA is not good for evaluating relievers. I’d like to read more on the topic.

        Vote -1 Vote +1

        • Bip says:

          It is not good for evaluating how good a pitcher a reliever is. WPA give relievers credit for things they don’t have direct control over, namely when they are put into a game. Two relievers might have equal skill but if one is used in the ninth and the other in the seventh, the one used in the ninth will have higher WPA. It doesn’t mean you should pay more for that one though. WAR will tell you more about which player is ‘better.’

          However, we are not trying to find out which one of these guys is better. This article is trying to illustrate that these guys were not worth the money being paid for them. However, while WAR may say a reliever is worth 1 win above a replacement player, based on the situations where he was used, he may have actually contributed 2 wins over a replacement reliever. 2 wins is 2 wins, and I would think you would want to pay for those 2 wins no matter how you get them.

          Darren is right, we would want the replacement-level version of WPA in actuality. Now that I think about it, we would actually want to just scale a player’s WAR based on the LI of his appearances — essentially a context-independent context-dependent measure — unless we buy into the idea of “clutch”, which WPA gives to the player.

          Vote -1 Vote +1

      • AC of DC says:

        Got no html skills to make this pretty, but to this notion:

        SORTED BY 2013 WPA (as listed on Fangraphs player pages)

        PLAYER (AAV in $M) WPA

        Grilli, J (3.375) 2.55
        Peralta, J (3.0) 0.97
        Soria, J (4.0) 0.70
        Burnett, S (4.0) 0.64
        Choate, R (2.5) 0.49
        Soriano, R (14.0) 0.24
        Adams, M (6.0) 0.10
        Broxton, J (7.0) -0.47
        Fujikawa, K(4.5) -0.61
        Affeldt, J (6.0) -1.87
        League, B (7.5) -1.94

        (Omitted Gorzelanny, T because starts/relief are not separated on the basic tables. For what it’s worth, he sits at -0.02 on the season combining all appearances.)

        Vote -1 Vote +1

        • Bip says:

          This makes sense. In general, these numbers are more extreme than their WAR numbers. Grilli is a legitimately valuable player, and if we used replacement level as the baseline, we’d find he becomes one of the most valuable players on the Pirates.

          On the other hand, League is a terrible reliever put into a situation where being terrible is most damaging to his team. His WPA should be worse than his WAR.

          Vote -1 Vote +1

  14. chief00 says:

    This article practically begs for some desperate, and perhaps heavily-medicated, GM to go out and sign Kevin Gregg to a 3-yr deal for $25MM to prove the point yet again. Long term, high AAV contracts for relievers are like pumping gas while smoking a good cigar. Sure it’s nice to spend too much dough on non-essentials, but that doesn’t make it smart.

    Vote -1 Vote +1

  15. Moves Like Munenori says:

    After shutting down Strasburg and that whole debacle, if Rizzo had let Soriano walk he may have been fired. He did let quite a few free agents go this past offseason.

    Vote -1 Vote +1

  16. Manifunk says:

    Everything old is new again, apparently. Dave is just lifting ideas from older articles: http://www.fangraphs.com/blogs/guest-post-why-not-to-overpay-relief-pitchers/

    Vote -1 Vote +1

  17. DodgersKingsoftheGalaxy says:

    Broxton pitched better? hahahahahahahhaahha

    Vote -1 Vote +1

  18. Ivan Grushenko says:

    The lesson I got from this article is that it’s better to waste $10M/year on 7 relievers and hope 4 are good, than to waste $10M/year on 1-2 proven closers.

    Vote -1 Vote +1

  19. Ty says:

    I have a question about the future performance of the whole of a team’s bullpen. What is a better indicator of future bullpen performance, ERA or xFIP? The reason why I ask this, is because a team’s bullpen will pitch about 500-700 innings in a season, and ERA is considered more predictive than xFIP in larger sample sizes.

    Vote -1 Vote +1

  20. BMarkham says:

    What do you think about Mujica, who will be a FA this year and will turn 30 at the start of next season? His FIP is more than a run higher than his ERA due to a ridiculous .224 BABIP.

    As a Cardinals fan I want to retain Mujica’s services but he’s really only supplied .7 fWAR so far this year. Some might say with Motte coming back next year we don’t need him. However, our bullpen Ace, Rosenthal, reportedly prefers to start so that may leave us down our 2 most important relievers next year.

    That being said, the Cardinals shouldn’t overpay. If Rosenthal leaves for the rotation then Wacha may take his place in the pen. And we could still have a bullpen of Motte, Wacha, Siegrist, Maness, Choate, and whoever else. Not the best bullpen in the majors by any means but seems pretty solid too. Retaining Mujica would give them one of the better pens in the league.

    If Mujica would take $10-14 million total for 2 years, I would take it. More years and money and I’m hesitant. Maybe a little more for a team-friendly option year. The problem is Mujica may demand much more than that with how many Saves he’s got this year. And with that BABIP ready to regress I wouldn’t want to be stuck holding the bill.

    Vote -1 Vote +1

  21. Scott Skink says:

    While the article is about the value of relief pitchers on multi-year contracts, is there any doubt that the least value provided from any reliever under any contract this year was that provided by Joel Hanrahan?

    For all the bad deals Neal Huntington has done, letting go of Hanrahan while getting Melancon (and other parts) in return may be the best single deal by any GM last offseason.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>