A Cy Young Reliever Should Be Almost Impossible
I love Zach Britton. I loved him even when he was a failure. I love his numbers, and I love his approach. I love trying to find other pitchers who might have some prayer of becoming the next version of Britton himself. It’s because of Zach Britton that I’m so interested in Blake Treinen! Britton has turned into something amazing, and this has been a particularly phenomenal season. I have a note written right here to compare Zach Britton and Mariano Rivera, and I’ve held off on pursuing that because it feels insane, even while at the same time it doesn’t. One pitch, here it comes. You’re a hitter, and you’re screwed.
Britton is as good as a closer gets, and he’s a huge, huge reason why the Orioles are where they are, even despite spending most of the year without a starting rotation. Britton might be the most valuable Oriole. He might be the most valuable…more than that. I just want to make absolutely sure that you understand how big a fan I am. It doesn’t get much better than watching Zach Britton pitch. He’s overwhelming in a different way from how, say, Aroldis Chapman is overwhelming. Much credit to him. I can’t support him as a Cy Young Award candidate.
This conversation has taken off lately, fueled in part by Britton, fueled in part by a relatively lackluster American League pitching class, and fueled in part by the post-deadline writer blues. As is always the case, because we’re talking about an award, this isn’t really all that important. And my own perspective is even less important, since I’m not a voter. I just feel like I should weigh in, since I might be the biggest Britton fan we have on staff. Britton is a lockdown closer. The way I see it, that doesn’t make him a strong Cy Young contender. A reliever being a strong Cy Young contender might be almost impossible.
To make the case for Britton, or any reliever, you have to talk about how important every inning is. It’s actually the whole argument, in a way — everything comes down to the context of the job. Britton has a gigantic WPA right now, because he pitches almost exclusively in high-leverage situations. Sure, he doesn’t have a starter’s workload, but when you fold leverage in, the gap narrows considerably. Britton has thrown 50 innings. Corey Kluber, say, has thrown 157. But Britton has an average plate-appearance leverage of 1.83, while Kluber comes in at 0.82. You could interpret that to mean Britton has thrown 91.5 “effective” innings, while Kluber’s at 128.7. Not so far now!
I don’t have a problem with the general thinking. I don’t have a problem with the math. I just disagree about what this is addressing. When you talk about situational context, you’re really talking about value. And you know how complicated that word can be, thanks to that other, even higher-profile award. There’s disagreement about the MVP every single season because there’s disagreement over how to define that critical middle word. If the Most Valuable Player Award were instead the Best Player Award, the conversations would be simpler. Importantly, to bring this back to the Cy Young — the word “value” doesn’t show up anywhere, in any form.
When you run an Internet search, you find that the Cy Young is supposed to go to the best pitcher. I asked Dave, since he was an NL Cy Young voter last season. He told me he was given zero criteria. He was just supposed to fill in some blanks. Perhaps we can get some guidance from the award itself? Here’s one from 2014:
That reads “The Outstanding American League Pitcher.” That’s “outstanding,” not “most valuable.” I couldn’t find a good picture of a 2015 award, but as blurry as this is, I make out the same words:
Again, presumably, “The Outstanding American League Pitcher.” What is interesting is that the award didn’t always read this way. Let’s go back a while:
“Most Valuable Pitcher.” But times have changed. The award itself has changed. I don’t know exactly why or exactly when, but I can’t imagine this was an accident. The physical award has gotten rid of the word “valuable,” and if that can serve as our guidance, then it should follow that for relievers, the bar is almost impossibly high.
“Outstanding” can be taken as a synonym of “best,” and while you could go with one literal interpretation to justify giving the NL award to someone like Alfredo Simon, we know that’s not how it’s meant. Now, absolutely, an effective reliever can be a league’s most outstanding per-inning pitcher. Zach Britton is probably the AL’s best per-inning pitcher. But when you talk about “best,” you strip away the context, the context that makes most of Britton’s whole case. What you’re left with are results and innings totals, and an understanding that in almost every single case, a relief pitcher is a failed starting pitcher.
It’s not that relievers are officially ineligible. But the award goes to the most outstanding pitcher, even though the category of “pitcher” has two subgroups. There are starters and there are relievers, and they have almost completely different roles. Starters throw five or six or seven innings. They pace themselves. They throw three or four pitches. They face guys multiple times. Relievers go at 100% for three hitters a game. Sometimes they’ll go up against five or six. You know all the differences. This isn’t a new conversation. You know that, statistically, just about every converted starter has more success in relief. Relieving is both super hard, and way, way easier.
On the position-player side, designated hitters get docked with a big penalty in the WAR formula, because it’s been determined they can’t help anywhere else. They don’t do a center fielder’s job. They don’t do a first baseman’s job. They’re assigned to do less. Relievers are a little like a pitching version of a DH, only instead of not doing a small fraction of the job of a starter, they don’t do the bulk of it. The Cy Young Award asks us to get starters and relievers on the same level. In theory we’d have to translate some numbers, either turning all starters into hypothetical relievers, or all relievers into hypothetical starters. Now this is just a thought exercise but if you were to imagine an excellent reliever as a starter, I doubt you’d have a No. 1. Because somewhere along the line the reliever was put in the bullpen in the first place.
I wouldn’t want to rule relievers out entirely. I’m open to other arguments. This particular argument actually just crystallized for me over the weekend, so it’s new to my brain. But I can’t get past two points. One, the Cy Young Award doesn’t say anything about “value.” And two, pretty much every single good reliever is a failed starter. Given those two statements, I don’t know how I could vote for Britton, even in a somewhat underwhelming AL candidate field. He’s as good as it gets in the ninth. The best pitchers get to hand the ball off, not receive it.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
1. Can’t we just make a new award for relievers? Keep the Cy Young and then create the Mo Rivera or some shit. These guys deserve to be recognized (and paid in arb).
2. Wouldn’t Sale or Syndergaard or Fernandez each be incredibly dominant relievers if they were asked to pitch one inning a game? The idea that the vast, vast majority of relievers are failed starters should be a major factor, as you said.
3. Looking at Rivera’s numbers, he only had K/9 >10 once (his first year as a reliever) and only had a gb rate higher than 60% once. Meanwhile, Britton is at 10.62 K/9 and 80% (!!!) gb rate this year (next highest in MLB is ya boy Blake Treinen at 65%).
There is an award for a relievers. Unfortunately, it doesn’t have quite the same history of the Cy Young. https://en.wikipedia.org/wiki/Major_League_Baseball_Reliever_of_the_Year_Award
To expand on Treggey’s point, naming it after Mariano Rivera may help give the award a bit more hype than it currently has.
The AL reliever award is named after Rivera. NL is named after Hoffman.
Mariano Rivera pitched in a much different offensive environment. A straight comparison of rates is misleading.
Something to consider: nearly every Cy Young candidate starter is utterly, HILARIOUSLY dominant coming out of the bullpen. This happens with reasonable frequency in desperate post-season situations, and the emergency-relief ace is generally very very very good. And that’s without adjusting their conditioning to account for the shorter outings.
Britton is amazing, but if Kimbrel didn’t win in 2012, Davis didn’t win in 2015, and Rivera didn’t win in ANY of his amazing years, Britton probably shouldn’t win it either.
I’ve never been a fan of people citing that “even Rivera didn’t win a Cy” as an argument that RPs shouldn’t be given much consideration for this. The thing about Rivera wasn’t that he was the absolute best in any given season, but nearly the best essentially EVERY season, for forever. Outside of his first full season when he was used very differently, I don’t think he ever had one of those absolutely ridiculous single years. He never should have won a Cy, not that that takes even the slightest away from his career as a whole.
Rivera had 11 seasons with ERA under .200 in an era when offense was crazy. I don’t think he should have won a Cy Young either, but most of his years were absolutely ridiculous, even if FIP doesn’t recognise it.
If Britton does win the Cy then, RA Dickey needs to give the 2012 Cy to Kimbrel
*cough* Kershaw *cough*
Interesting that one of the commenters to August’s post on the same subject made the same comparison of relievers to DHs that you have, and got what must be a near record of negative votes for it.
Not that I disagree. Actually, an interesting debate might be, which is more likely, or should be more likely, a DH winning MVP, or a reliever winning Cy Young? Papi has slipped a little, but for the first half he was hitting at a level close to what Harper did last year. Would a full season like that from a DH make a more reasonable argument for MVP than what Britton has done this year for Cy Young?
I’m pretty sure the downvotes had more to do with that poster describing Mr. Fagerstrom’s column as “a trash idea.“
It really is amazing how many people think that how you say something has nothing to do with his people receive it.
I don’t think relievers should win the Cy Young, but I don’t have a problem with a DH being MVP if he’s a vastly superior hitter than anyone else.
Comparing WPA between pitchers and relievers doesn’t feel fair to me as a way to gauge value. A quick rant:
(1) It feels too much like it’s partly a team stat, like pitcher wins, or RBI, in the sense that the reliever is getting credit for making the most of the opportunities he’s given. Nothing wrong with that! But because a reliever isn’t responsible for the situations that have been available for him, it’s of limited value when comparing how good the players are. For example, in the right circumstances, a pitch-hitter like Matt Stairs (great hitter, love Matt Stairs, but a limited player) could amass a lot of WPA. But that reflects not only his performance but the situations that the team had found itself it in. I think a reliever is somewhat analogous to that. If Matt Stairs threw up a bunch of WPA because of some great timing on his pitch-hitting, would we put him in the MVP discussion? I don’t think so. So I don’t think WPA is as useful a way to evaluate players’ ‘value’ as players.
(2) Boy howdy does comparing the WPA of starters and relievers get weird in a hurry. Jason Grilli has amassed more WPA for the Jays (1.20) than Noah Syndergaard has for the Mets (0.97) this year. Has Grilli’s 25.2 IP been 20% more valuable than Thor’s 134.1? My intuition is no one would take that position. Grilli’s been great for the Jays, just amazing, but Thor was amazing too, in greater quantity. Comparing WPA between starters and relievers punishes starters for run support! It punishes them for pitching early in games! It rewards the relievers for being better than the other relievers on the same team so they get the high-leverage spots!
End of rant. Sorry.
Oh, sorry – to be clear, great article, I totally agree.
I would take the Britton arguments more seriously if he was actually leading all RP in WAR… but he isn’t even that.
That’s because fWAR does not take into account GB% (other than lack of HRs). Britton leads rWAR with 3.1. Behind him are team mate Brad Brach at 2.8, Andrew Miller at 2.5, and Addison Reed at 2.2 That’s a pretty sizeable lead.
True but Dellin has a significantly lower FIP and xFIP. Miller beats him in xFIP as well by a good margin. If we’re talking true talent or who is really the best pitcher in 2016 does Britton’s GB% make up for the all the extra strikeouts of guys like those 2? I’m sure Britton gets weak contact but his .212 BABIP is insane. It’s an interesting debate.
According to Fangraphs, groundballs result in a .239 average. And seeing as how Britton is one of the best and almost exclusively pitches ground balls, it’s not that absurd his BABIP is that low.
Almost every dominant reliever is a failed starter, but certainly not every starter makes a dominant reliever. You cannot dismiss the fact that just being able to pitch does not mean you will do better coming out of the bullpen. The fact that Britton failed as a starter should not be held against him.
Well I think it definitely should. It shouldn’t start and end the conversation or anything, but he’s not a reliever because he wanted to be, or because it was the team’s desired outcome.
I don’t think referencing the term “outstanding” really does the work to answer the question you pose. Outstanding is defined by various dictionaries roughly as “exceptionally good.” What makes a pitcher a good pitcher? One who has good raw pitching ability? That tends to support your point. But also plausible is to say that a pitcher is good when he accomplishes the purpose of a pitcher as well as possible, which is to say he provides value for his team. I’m inclined to think that your interpretation fits the definition better, I’m only maybe 60-70% sure of that.
Are we sure Britton is the same pitcher now that he was when he failed at starting?
Outside of the obvious velo increase -maybe he has changed? How would he fair as a starter today?
He still has 3 pitches he uses, and it looks as if he dropped an eh 4th pitch. His velo increased 5 mph the year he moved to the pen, probably mostly related to the switch, but maybe not..
Britton has been better each year since he switched, so it’s not all just because of the switch. He’s added an extra MPH to his velo since his 2014 BP debut . His o-swing rate has increased, his groundball rate has skyrocketed during that time.
Maybe his sinker is good enough to go 3 times through a lineup? We don’t know, because Britton isn’t starting.
As an O’s fan, i still don’t think he’d make it as a starter. He still has games where he’s pretty wild, and if he wasn’t able to put some extra heat on the pitches i don’t think he’d be effective those days.
I don’t think he could keep it up over more than a couple innings on his bad days.
This is an “eye test” statement. His BB ratio is perfectly fine. Maybe you perceive he doesn’t have command, but his results say otherwise.
Of course it is the total lack of ‘outstanding’ starters that’s opened the door to this Britton chitchat in the first place. Britton is standing out in his role, while the starters are, say, huddling together. Strict literalism here would almost demand a vote for Britton.
Miguel Cabrera didn’t stick at 2B in Utica in 2000. Do we now say he’s “only” a failed 2B?
Not that I think Britton should get the CY, but guys eventually find their place, and if they excel in that, I don’t think it’s fair to call them a failure at earlier roles. I agree with the rest of your take on value, but not about factoring in this arbitrary characterization.
Mick, maybe saying he is “failed” is a bit harsh, but yeah, when we evaluate position players we consider their defensive value alongside their offensive value. Part of defensive value is the position the player fields. And Jeff’s article flags this point when he discusses the DH/reliever comp. If Miggy put up the same offensive numbers from 2B that he did from 1B/3B/DH then he would have contributed even more value. Now, Miggy is a HOFer in any event, but if he had produced the same level of offense from 2B, then we’d be discussing him as potentially the best 2B of all time right now.
Look at Chase Utley’s career WAR, and think about how much different it would look if he put up those same offensive numbers from 1B.
Comparing a position switch to a move to the bullpen isn’t accurate because the majority of the value of a guy like miggy comes from his bat…..a more accurate comparison would be a hitter who was initially in his team’s lineup but is now used as a pinch hitter, in which case you can obviously agree that the pinch hitter is not nearly as valuable as an everyday player, even if his at-bats come in high-leverage situations
You can’t compare switching positions to moving to the bullpen. Switching positions requires learning something completely different. Moving to the bullpen only changes how many batters you face. The pitching aspect is still almost entirely the same.
This argument doesn’t do what it claims to do, and the the reason really is that the all-too-common relievers are like designated hitters analogy is specifically inaccurate. We can see why easily when we quote the actually correct description of what starters and relievers are asked to do that occurs in the paragraph before the DH analogy comes up: “Starters throw five or six or seven innings. They pace themselves. They throw three or four pitches. They face guys multiple times. Relievers go at 100% for three hitters a game. Sometimes they’ll go up against five or six.” This is not the difference between fielders and DHs. It’s the difference between distance runners and sprinters.
Now, you are welcome to say to me “there should not be an award that is common to marathon champions and 100 meter dash champions.” For the most part, there isn’t. What you are not welcome to say to me is that marathon champions are more “outstanding” runners than 100 meter dash champions. (And starting pitchers aren’t even marathoners, least of all these days; they run the 1500. Maybe a complete game is the 3000.)
I see nothing strange about the idea that Britton is this year’s most “outstanding” AL pitcher. And I haven’t the slightest idea whether, if he is that thing, he “should” win the Cy Young Award.
The problem with a distance runner vs sprinter analogy is that there isn’t a history of mediocre distance runners switching over to sprinting to find overwhelming success. (Correct me if I’m wrong there, I don’t know track and field athletes particularly well) While you’re right in noting that the fielder/DH argument isn’t completely apt, it’s an understandable analogy for a baseball perspective. A DH is “shielded” away from his inherent weakness of being unable to field a position as a reliever is shielded away from his inherent weakness of being unable to get batters out multiple times through the lineup.
The fact that players move from starter to reliever in baseball emerges from the logic of baseball, but doesn’t necessarily tell us anything about what people are as athletes. All young people who show pitching aptitude start as starters; there is no other direction for them to move. Though I’d love to have a high school track coach check in here with real information, I presume that there isn’t anything about high school track that specifically pushes runners to do one thing first and then switch to an apparently less valuable race later. If there was a logic to stopping a young person capable of throwing 10 pitches at 100 miles an hour from trying to learn pacing and changing of speeds, we might learn something about how to train pure relievers as we train pure sprinters. We also might not. I have no problem throwing my hands up on that one. My only point is that it proves nothing about relievers that failed starters become relievers if 100% of potential major leaguers are encouraged to be starters first.
Distance running and sprinting require entirely different skill sets and training, even different groups of muscles. They are essentially two different sports, like sprinting and swimming. Starters and relievers still essentially do the same things.
To what extent is simply being a starting pitcher more “outstanding” than a reliever or failed starter like Britton? Would you consider Chris Tillman more outstanding than Britton since he is a starting pitcher? How about Jered Weaver? We can agree he is not a good starting pitcher, but is he more outstanding than Britton with his 130 innings?
Since becoming a full-time reliever on July 8th of 2015, Joe Blanton has pitched to a 2.38 ERA and 2.99 FIP. In other words, I agree with you Jeff.
As others have said, Britton is essentially a bench player, and if he was good enough to start he would be starting. But he failed as a starter. That should be enough to eliminate his from candidacy in the mind of voters.
And look at all the dominant relievers who were failed starters. Mariano Rivera? Failed starter. Joe Nathan? Failed starter. Dellin Betances? Failed starter. Wade Davis? Failed starter. Lets stop this nonsense about relievers winning the Cy Young.
The fact that relievers are, on the whole, failed starters is completely irrelevant to whether they should win the Cy Young Award. That’s like saying first basemen should never win the MVP because they are failed shortstops. Or Tony LaRussa should not have made the Hall of Fame because he was a failed player.
The problem with relievers in the modern game is they simply do not pitch enough innings. It is difficult to make 60 innings as valuable as 200 innings regardless of leverage. It is made worse by teams insistent on using their ace reliever in save situations regardless of leverage. If Mariano Rivera had been used as he was in 1996 and regularly pitched 100+ innings of high leverage, high quality work, he probably have won a Cy Young and deserved it.
No, it’s more like saying a bench player and pinch hitter should never win the MVP. Even if that player hit .400/.500/.600 in 100 PA over a season they shouldn’t be considered for the MVP over guys who had 6 times that many PA.
Except the example given is 100+ innings for a reliever, which is closer to a starter’s output than the bench player you introduce to invalidate the point.
The way relievers are used now, yes, it is like a bench player winning the MVP. A 450 season with 10 homers over 120 PAs is not MVP caliber unless the league essentially has no position player stars and all the starters are worth 2 WAR and less. However, a bench player that had 300 PAs and batted 800 with 30 homers would be an MVP candidate. (Yes, I do realize how unlikely that is.) That’s the point I am making. The problem is not his position is relief pitcher. The problem is pitching 60 innings.
Mariano Rivera in 1996 was worth 5.0 bWAR/4.3 fWAR. He had better years with his ERA, relative to the league. He was technically not even the relief ace that year on his own team. During his career, the bWAR leader in the AL was as low as 6.5, 6.2 in the NL. There have been rare seasons where the best pitcher by bWAR was in the 5s. (At the moment, Kershaw has the best bWAR in the NL at 4.8 though someone should pass that eventually.) It is very conceivable that a relief pitcher used for 2 innings at a time in consistently high leverage situations and still retain the same effectiveness – the reliever should still only be going through the line-up once – could be among the best WAR pitchers in any given season. This is not to say that every relief ace could be converted to this mode of pitching and still be great – I’m sure there are really guys who are only 1 inning pitchers – but some pitchers certainly can do that. And an elite reliever who could do that would be very valuable indeed.
Suppose you are the Dodgers, so Kershaw is your ace. Lucky you! But another team has a reliever, lets call him Mariano Chapman, who never gives up a run. Ever. But he can only pitch for a limited number of innings each year. How many innings would that have to be for you to prefer Mariano Chapman?
Its a question of where you draw the line, not whether or not the line exists.
That was one extreme case, but there is a continuum of possibilities. What about the old school relievers who would pitch multiple innings and go to 100-130 IP on the season. Instead of an ace with 2.50 RA9 in 200 innings, you get 1.50 RA9 in 120 innings (of higher leverage). Which one is more valuable?
We dont yet understand where that tradeoff is, so its a great debate. Perhaps someone wants to runs some Monte Carlo simulations…
I really hope Mariano Chapman throws a 105 mph cutter.
There are probably boat loads of former starting pitchers that failed as relievers. They are a bit less visible though because they are all out of baseball.
Also, Johan Santana is a failed relief pitcher that became a HOF level starter.
If you take out the fact he used PEDs do you think Eric Gagne deserved the Cy Young in 2003?
I think that WPA fails as an advanced metric because it is too dependent on team context.
Reliever usage is pretty uniform, and the best ones tend to pitch only the 8th or 9th innings when a team has a lead of exactly 1-3 runs.
Granted, a run allowed by a reliever is more likely to influence the outcome of the game, but some of that is due to the overall makeup of the team. A team like the 2016 Cubs, with both a good offense and a good starting rotation, seem more likely, at least on the surface, to have more 2-3 run leads going into the 8th or 9th inning that a team like the 2016 Orioles, who have a good offense but a horrible starting rotation.
Some of this highlights the fallacy of the save rule – a 3 run lead in the 9th isn’t a “close game”, and WPA adequately captures this. Obviously, a 1 run lead in the 9th is, but Britton has nothing to do with the fact that his team’s rotation stinks.
Over the long run, this should standardize itself. For example, over the course of Mariano Rivera’s career, one presumes that he pitched in a roughly equal number of 1 run, 2 run, and 3 run games. That said, his career WPA outpaces his career WAR by a ratio of about 55/40, so perhaps there is an argument to be made that WAR undervalues a reliever a bit, but certainly not at a level that equal’s Britton’s WPA to WAR ratio this year.
There’s no getting around the fact that advance metrics try to evaluate players on both qualitative and quantitative aspects, and because of this it’s just extraordinarily difficult for a top line reliever to be worth more than a top line starter, when the starter is pitching 150 more innings.
I know analogy’s from other sports do not make the strongest augments, but I can not help myself. Comparing starting pitchers and relief pitchers is like comparing the athletes that run the 100 meter dash with ones who run longer races. Now imagine relay race where one athlete runs 600 meters and three of his teammates run 100 meters for an odd 4 by 900 relay. In that sport it is almost unimaginable that Usain Bolt would be the best runner.