More Relief
Mariano Rivera is perhaps the greatest closer of all time. In 2008, Rivera posted a 2.03 FIP, the best of his career as a closer, while recording 70.2 innings pitched. Given such, you would expect Rivera to be extremely valuable amongst all pitchers, not simply relievers. As it turns out, Rivera’s value was 4.2 wins, or the wins equivalent to James Shields and Aubrey Huff.
Carlos Marmol acted as the Cubs eighth inning presence and was worth all of 1.5 wins. The currently unemployed Garret Anderson was also worth 1.5 wins. Ryan Madson just signed a juicy extension, but in 2008, he was equal to Edwin Jackson and Gregor Blanco. Of the 332 pitchers who completed at least 10 innings in relief, only 4 were valued higher than 3 wins (Rivera, Papelbon, Chamberlain, Fuentes), 17 finished with greater than 2 wins, and 55 with at least one win, leaving 277 relievers worth a win or negative value.
Of course that’s with our innings limit set relatively low, and bumping the threshold up to 50 innings results in 135 qualifiers. From there, we have the following results:
4+ wins: 1 pitchers
3+ wins: 3 pitchers
2+ wins: 14 pitchers
1+ wins: 47 pitchers
0+ wins:115 pitchers
Negative value: 20 pitchers
Or, in percentage form:
4+ wins: 0.74%
3+ wins: 2.22%
2+ wins: 10.4%
1+ wins: 34.8%
0+ wins: 85.2%
Negative value: 14.8%
Yes, you are reading that right; more relievers with 50+ innings had negative value than those who had 2+ wins, or “average” value from starting pitchers and batters. Also somewhat unbelievably, the 14 pitchers worth 2+ wins only averaged six additional innings than the 20 with negative value. Of course, there’s also the matter of which relievers would pitch tons of innings, seemingly either the very best or the worst would be turned to often. This is to say, set-up men and closers and mop-up men.
Stating that middle relievers are overrated is hardly groundbreaking, but I’m going to state it anyways: middle relievers are overrated.
It’s almost like bullpens in general are overrated. Give me a stud reliever and I’m happy. I’ll invest the rest of my money and talent in position players and starters.
One reason bullpens get too much attention is that given the higher LIs they accumulate, over- and under-performance gets magnified. A 3.50 FIP true-talent bullpen that manages to post a 3.00 ERA is going to be more valuable than just those .50 points of ERA. And a bullpen that underperforms it’s true talent will cost it’s teams more games. That under- and over-performance isn’t due to skill, though, and next year could flip flop in the other direction. There’s a lot of luck involved in a bullpen’s performance and people tend not to separate the skill from the luck very well.
Great writeup. Excellent analysis, and although it merely confirms what we already knew (as you said, middle-relievers being overrated), it is certainly interesting seeing that quantitatively laid out.
Loved this piece, RJ. I think it’s right on the money. Bullpens — aside from the +3 win relievers like you listed above — just don’t add as much value to a team as most assume. Especially when you consider how much the media has inflated the concept of the Prove Closer(tm).
Are Chamberlain’s wins boosted by his time as a starter or did you solely look at his innings of relief pitching?
Thanks for this article though. As a Yankee fan it gives me one more article to throw in the face of the “Joba to the bullpen” crowd.
Relief work only. That’s also why he didn’t qualify once we raised the innings threshold.
Does the FIP/wins measure incorporate leverage?
Yep.
Im really having a hard time conceptualizing that IP have so much value, even at a very high FIP. The replacement level has to be off… finding a guy who can throw 170 IP of 6.35 FIP is not that difficult. There are loads of these guys hanging around. How can that be worth so much (In Robertson’s case, 1.3 wins/5.8M)? How can this be a desirable result from your pitching win value system? In reality, Detroit could have replaced Robertson with pretty much anyone. Replacement level has to be higher, by definition.
Nate Robertson didn’t have a 6.35 FIP. He had a 4.99 FIP.
Replacement level isn’t off. You’re evaluating him by his ERA.
Ahh… thank you. That makes sense.
So, you’re saying that a 3 years 19 million dollar contract was not a good deal for Danys Baez in ’07?
Maybe this was covered elsewhere, but I don’t understand. Are you using WAR? It would be helpful if you said so. What is it about the approach to WAR that makes relievers less valuable?
And in what way are middle relievers overrated? I don’t see any evidence for that in your post. Overrated compared to what?
I don’t mean to be picky, but I’d just like to better understand what you’re saying and why.
“middle relievers are overrated”
Tell that to the 2007-2008 New York Mets…
I see the point though.
I think that statement was not so much based on their importance, as much as it was based on the fact they get paid a fairly significant amount to accomplish very little (and often times prove to be a detriment to the team).
If anything, the Mets were the poster children of MRs being overrated. That statement, semantically, can really be looked at either way. Perhaps just the greatest ones are underrated given so few actually help their team.
That said, I wonder if a calculation adjustment needs to be made for MRs. The reason being, if the vast majority (and in this case, the vast, vast majority) hinder a team’s chances of winning a ballgame over the course of a season, perhaps things needs to be shifted a bit in regards to MRs and wins. I suspect that MRs today aren’t all that different than those of 10 years ago, so perhaps a MR that simply “loses” you -1.0 games per season ends up being pretty damn good relatively. I mean, your worth on the market is based on those around you for the most part. If I know I’m going to lose games with my MRs, wouldn’t you want to limit the total loss as best possible? Sounds simplistic, but I think the definition of “overrated” perhaps needs to experience a relative shift due to what is being considered here.
And the 2009 Mets are going to be dramatic evidence that overpaying for free agent relievers is one of the least efficient uses of payroll dollars. However, given that they have the revenue available thanks to the new stadium, and in lieu of other glaring weaknesses for a team that’s been agonizingly close to the postseason twice now, they can justify being inefficient. I’m not sure they can justify being quite as extravagant as they were — either the K-Rod or the Putz deal by itself might be defensible, but the two together is kind of ridiculous — but with the sour taste of the past couple of seasons still lingering on tongues amplified by the NY media environment (not to mention keeping up with those thrifty guys in the Bronx), Minaya probably had to go big or go away.
I am sure each team would love to have a stud reliever at each spot in there pen, but allocating the necessary payroll to do so obviously would be unrealistic. Except maybe to the Yanks…
What this highlights is not just the marginal value of acquiring “stud” relievers, which are next to impossible to find, but the marginal value of acquiring decent starters, guys who will get you to the 8th inning so you don’t have to rely on so many BP innings.
I find it amazing that in 2007 only one reliever was worth more than 4 wins – and that the reliever was Rafael Betancourt
Mariano is really amazing. It’s not about 1 season. There have been many relievers who have had excellent seasons. Mo is often not the best reliever in any given season. But every year, he’s one of the best. I will miss that terribly when it’s over.