Platooning Closers: Good Idea or Great Idea?
Yesterday, Atlanta Braves manager Fredi Gonzalez suggested that fireballer Craig Kimbrel and left-handed Jonny Venters might share closing duties this season barring a surprising return by Billy Wagner from presumed retirement. The Braves certainly aren’t unfamiliar with this strategy: Bobby Cox used both the right-handed Rafael Soriano and left-handed Mike Gonzalez in save opportunities in 2009 before Cox went primarily with Soriano. So while the 22-year-old Kimbrel was supposedly slated to be the team’s primary closer after quite a September run and a brief role in 9th inning situations, this option reveals an adaptability that the Braves management is willing to take.
Since the ‘save’ became an official Major League Baseball statistic in 1969, teams and fans have overused the term, misguidedly limiting a team’s best reliever into a closer’s role. Not to say that it isn’t beneficial to have some sort of consistency, but when you save your best reliever for last and don’t employ the flexibility to bring him out during high-leverage inning situations that often occur in the 7th or 8th innings, you do your opponents a service by not optimizing your reliever usage.
So I like this idea. No, I love this idea. This doesn’t undercut Kimbrel’s performance, though. He definitely has the stuff to close and has shown the ability to dominate. His mid-high 90s fastball is characterized by its movement, while his slider, his other plus-plus pitch, gets swings and misses. And the results have been pretty spectacular. Kimbrel struck out the side four times out of 21 appearances in 2010, stranded 92% of runners, and recorded a 0.44 ERA with a 17.42(!) K/9. Turns out that’s the highest single-season strikeout rate in HISTORY for all pitchers with at least 20 innings pitched (h/t to Dark Overlord Appelman for adding single seasons to FanGraphs’ Leaderboards today). Now that’s impressive.
What splitting the closer’s job with Venters may do for Kimbrel is allow him to work on his major control issues and relieve some of the pressure with giving a young player the closer’s job early for an extended period of time. Being able to consistently throw strikes like in September and the playoffs has not been, in particular, the norm for Kimbrel, not with the 6.97 BB/9 in the majors and the 5.00+ BB/9 rates posted in Double-A and Triple-A. And after quite a dominating late-season run, it’s still worth noting that Kimbrel has not thrown more than 25 innings in the MLB including the postseason. It takes time for young fireballers to be groomed, especially ones with such control problems as Kimbrel’s.
Venters, on the other hand, pitched for most of the 2010 season, amassing 83 innings of 1.95 ERA ball and 10.08 K/9, 4.23 BB/9, and 0.11 HR/9 rates, also very impressive but also with control issues. Venters actually didn’t have a great K rate throughout the minors, so I’m not so sure he’ll continue to post such numbers. The hope is that he can keep his very good sinking fastball low and continue to get both whiffs and groundouts, but he’ll also have to work on control issues in light of all the zipping movement on his fastball. The lefty has more experience and very good numbers against left-handed batters (14.79 K/9, 3.21 BB/9, 1.97 FIP) that he’ll be preferred over Kimbrel in 9th-inning situations if you have hitters like Chase Utley and Ryan Howard coming up. He’s also not too shabby at all against right-handed batters and looks like is capable as one of the few left-handed closers in the game.
Saves apologists will cite the Red Sox closer-by-committee experiment in 2003 as evidence that such an approach does more harm than good, but the Sox simply didn’t have the talented relievers back then to make it work. In fact, earlier this year, the over-reliance on the ‘saving closers for the 9th inning’ fallacy cost Ron Washington and the Texas Rangers Game One of the ALCS against the New York Yankees when Washington marched out five pitchers to blow a five-run lead, none of them being the Rangers’ best reliever, Neftali Feliz. A no-outs, bases-loaded, three-run lead in the 8th inning should be a situation in which you bring out your best stuff. Instead, Washington put his faith in Darren O’Day.
So if the goal of a Major League manager is to improve the chances of winning a ball game, playing matchups and bringing the best relievers in the most pressured situations improves your chances. Keeping the game decisions and a variety of choices in your hands rather than being chained to the myth that a closer can pitch if and only if it is the 9th inning with at most a three-run lead can do a lot for the flexibility of the bullpen. Both pitchers need to work on control issues, with Kimbrel dominating mostly with strikeouts and Venters drawing whiffs from left-handed batters and groundouts from right-handed batters. Heading into the 8th inning of any game, with both ready either to set up or to close the game, Gonzalez can elect to play the matchups and situation, factoring in how much rest they’ve had and managing the bullpen in his hands.
At the same time, while I like this idea to start out the season, eventually Gonzalez and the Braves will concede and elect a closer out of the two. But by then, both pitchers will have worked on their control issues with a little bit of competition between them and a little bit less in-game pressure situations. So even if the Braves eventually do pick a closer, they’ll have done so given more experience and more innings pitched by both young relievers.
Albert Lyu (@thinkbluecrew, LinkedIn) is a graduate student at the Georgia Institute of Technology, but will always root for his beloved Northwestern Wildcats. Feel free to email him with any comments or suggestions.
How did you write this entire article without even mentioning that the Yankee’s currently have 2 closers? Hopefully they use Mo as the regular closer and Soriano as the shut down guy (comes in any inning after 6 with the game on the line to stop runs from crossing the plate.
Because Mo is the only closer – everyone else in the world is a set up man.
Because the Yankees have 1 closer, and 1 guy who used to close.
Given the way most managers utilize their bullpen, Soriano will likely be the 8th inning guy no matter what. I agree that Soriano should be the shutdown guy sometime in the late innings as long as the logistics of warming him up in the 7th and keeping him warm doesn’t affect his durability. I know teams USED to do that with relievers all the time, so i don’t know why it would be that big of a deal, but they would have to get him warmed up in advance, otherwise, that high-leverage situation can creep up on you quicker than you can get your guy warmed up.
big question is… will soriano generate more value than Rivera as an 8th inning/shut down guy?
I guess that’s a big part of the debate…and if Soriano and Mo both have good seasons, we will have a good example to look at this
Venters was primarily a starter throughout the minor leagues, which might help to explain the disparity in K-rate b/w the majors and minors.
He also seems to have gained a fair bit of velocity between the minors and the majors which I tend to attribute to the change to the bullpen. He sat 93-96 this year (fangraphs has average at 94.6) and the minor league scouting report (or at least the one I have in the back of my head for all Braves prospects…) had him at something more like 90-92.
Adding 3 mph will do okay for your K rate, too.
That’s a very good point, I should have mentioned that, thanks. Being able to exert more into a fastball that relies on sinking movement will change a strikeout pitcher’s rate stats when he converts from a starter to a reliever, not the least of which is additional up-ticks in speed and movement.
The added velocity makes Venters’s fastball simply disgusting. He was a fairly undistinguished starter in the minors, but the added velocity has made him a true ace reliever. A pitcher with a mid-90s sinker — like a pitcher with a mid-90s cutter — really doesn’t need many other secondary pitches.
I have always liked the idea of two Left/Right closers. It adds a ton of flexibility and allows better match ups for the pitching team when leverage is the highest. This allows the following strategy at times: before the 8th inning, the manager can foresee when the “heart” of the other team’s batting order is likely to come up (8th or 9th inning), evaluate lefty-righty issues, and use of the two closers in the 8th inning. I think the main impediment is that veteran relief pitchers have been too used to the closer designation and will resist a mix and match approach. If a team had two young pitchers who were taking on the closer role for the first time, maybe it would be easier. The use of a Left/Right combo for closer was common place in the 70’s and 80’s. Even into the 90’s, this was a common approach if a team had two good relievers of the left and right hand.
I’ll throw this added benefit out there, too:
In a market where saves are the currency of closers, by keeping both Venters’ and Kimbrel’s saves number down, neither one will be paid a true closer’s salary.
This is good for the Braves not only because closers are paid quite handsomely, but also because relief pitchers are (in my opinion, and apparently Billy Beane’s opinion as well), quite undervalued.
Wait… relievers are undervalued? How is it that the best-paid players (in terms of free agent $$$ relative to workload and WAR) are “quite undervalued”?
I’m pretty sure he meant relievers who don’t record saves. Whether or not it’s true, it’s plausible.
The way major league managers use their bullpens is rediculous. The closer (presumably the best reliever) should be used in the highest leverage situation, not necessarily at the end of the game. For instance, if your team is up by one going into the eighth inning with the top of the lineup coming up that is when the best relief pitcher should be used not the ninth inning. It would probably be an even better idea to simply not have a designated closer and simply rely on matchups rather than waiting to use your best pitcher until the last inning.
Well, no. You don’t know what a game’s highest leverage situation is until after the game’s over, evidenced by your example being [b]wrong[/b] – if your team is up by one going into the eighth, you should not use your best reliever, because, best case, your best reliever preserves the lead and you’re faced with a higher leverage situation in the ninth, since the score is the same. You can make educated guesses, but you can’t know for sure. For example, would you use your best reliever in the 4th if you’re up by one and it’s bases loaded, no outs? How about if it’s the 5th? 7th? Obviously it’s not cut-and-dried.
I have often felt that teams should put their second best reliever at closer. The best should setup. If you look at teams of the past. Teams with truly dominant setup men seem to always do very well (Wetland and Rivera). Meanwhile, dominant closers often play for bad teams such as the Royals because the setup situation is bad.
I would say its more likely that only good teams keep dominant set up men – if your team isn’t in contention at the deadline you will probably trade your dominant set up man (or trade your closer and promote your set up man).
“Meanwhile, dominant closers often play for bad teams such as the Royals because the setup situation is bad.”
I don’t understand how that makes any sense whatsoever. Just because Soria plays on a crappy team, doesn’t mean you say it happens “often”.
Selection bias. Teams who can afford (in terms of opportunity cost) to have “truly dominant setup men” generally have very talented rosters – and, at the very least, have one other truly dominant reliever.
Flashes of Ryan Madson come to mind
why does gonzalez have to pick one?
Funny, I stated “I don’t just like this idea, I love this idea” on my twitter right as I read it as well. Great minds.
Whitey Herzog did this 30 years ago.
And really did it, switching the RHR to LF and bringing in a lefty for one batter, then moving the RHR back to the mound.
Not that anyone since then has had the nerve to stick a RP in the OF for one batter.
I’ve seen LaRussa and Pinella move a reliever to the OF for a platoon matchup before.
It’s rarer these days, but it happens. It should happen more in the AL, as well
Bobby Cox did this in 2008 ( I believe) with Chris Resop. I remember it didn’t work out very well, but it was still very interesting.
I remember the Mets did this occasionally in the mid-1980s with McDowell and Orosco. Both were halfway decent athletes, so even if the ball was actually hit their way, they had a shot of catching it. But i don’t recall ever seeing either of them having to make a play in the field.
I could have sworn I saw another team do this in the past couple of years (other than the Braves example above), but I don’t recall the details at the moment.
Great Article, would really like to see the O’s try the this year. Koji and Gregg (and Maybe Gonzalez?)
Has anyone ever done a study that shows if a team used its closer “best reliever” in 70 games, but maximized their use by only using them in their teams highest 70 leverage innings during the year as opposed to who was used alteratively, to study the ultimate impact on the difference in runs allowed?
Nick,
I believe something similar what you’re asking was covered in “Baseball Between the Numbers.”
There is something to be said for having set roles too. Guys like to have a routine and those roles help with that. You can’t ignore the human element when constructing a roster. That’s one of the weaknesses for a closer by committee approach.
That’s not to say its not workable as I agree that an artificial statistic (The Save) has had a somewhat negative impact on game management.
I know the A’s aren’t claiming this, but they really do have this system already in place. They just signed Fuentes (a lefty closer) to be “just another bullpen arm.” It’s reasonable to assume that Fuentes will be closing some games when lefties happen to be coming up in the ninth.
With some success, Davey Johnson used Jesse Orosco and Roger McDowell as platoon closers in the 80s.
The Mets should platoon that fuck shithead K-Rod so that he does not hit the easily attainable option contained in his 2011 contract.
If he hits it, they are screwed for 2012 – K-Rod will get paid $17.5 million.
After reading this whole thread,
one thought that is not included,
is that pitchers all have different psyches.
Too often, a guy gets moved to the 9th
because he’s pitched well as a middle reliever,
but just doesn’t have the mind a 9th inning guy needs.
I agree with the lefty righty matchup, and I actually think Venters and Kimbrell both have the mental stuff to close games and pitch in high leverage situations. I just think guys like Farnsworth should never be closing
Other than LaTroy Hawkins, who else are you thinking of who “just doesn’t have the mind a 9th inning guy needs”? I can’t think of too many guys who thrived in a setup role who just busted in the 9th.
I don’t like the platoon idea. It hurts fantasy stats!