Stop me if you’ve heard this before: the Mets bullpen is bad. The pen’s 5.28 ERA is second worst in all of Major League Baseball. They also rank in the lower half of the league in HR/9 and in the lower quarter in K/9.
While Terry Collins deserves his fair share of the blame for his utterly confusing bullpen management, the modern “norms” of bullpen usage deserve to shoulder at least some of the burden.
What norms do I speak of? The closer’s role and the silly idea that is engrained in the minds of most baseball folks that he is to be used only in ninth inning save situations.
There is perhaps nothing more detrimental to the functionality of a bullpen than taking your best reliever and pigeon-holing him into only pitching in a certain inning, regardless of game situation.
Before I go and introduce my solution, here’s a quick introduction to the Leverage Index. The Leverage Index calculates the, well, leverage of the game situation when a reliever is used. The higher the Leverage, the more important the situation is in the game. Simple stuff.
That being said, my solution to the Mets bullpen woes is to abolish the closer role altogether. You simply take the best reliever on the team, in the Mets’ case Bobby Parnell, and make him the ace reliever. This means that you bring him in for the highest Leverage situations, whenever they may occur. A lot of times that will be in the ninth inning, sometimes it will be the seventh or eighth.
Jonah Keri explains uses numbers and research to explain this concept very well in his book, Baseball Between the Numbers (page 69):
“Over the span of 2000 to 2004, the median maximum Leverage was 1.66. Once the game situation has a Leverage exceeding 1.66 – meaning a situation where allowing a run to score has a 66 percent more impact on the likelihood of winning than it did at the start of the game – it becomes one where we should consider bringing the ace [reliever] into the game.”[1]
Reconstructing the Mets bullpen to operate without a set closer might be the best option for the team to get the most out of the limited abilities of this year’s version of the ‘pen.
Collins would have more flexibility when bringing in a reliever, and, more importantly it would allow the team’s best reliever to pitch in the highest leverage situations, helping the team to win more games.
Now this is hardly a new theory; in the 70’s and 80’s the ‘ace reliever’ was used for multiple innings and in these high leverage situations. Mike Marshall, Bruce Sutter and Dan Quisenberry are examples of pitchers used in this fashion.
It was not until the 90’s when pitchers like Trevor Hoffman, Billy Wagner and Mariano Rivera came along that the closer’s role changed to what we know it to be today.
But has this adherence to silly and arbitrary labels really cost the Mets in terms of wins? Unfortunately we don’t have to go back too far to get the answer.
Let’s take the April 25 game against the Dodgers. Please.
As you may recall, the Mets lost that game 3-2 after Scott Rice surrendered two runs in a 1-1 game in the ninth inning (Ike Davis would homer in the bottom of the ninth for the other Met run). After the Dodgers had taken a 3-1 lead, Terry Collins went to the bullpen to bring in Bobby Parnell to mop up the mess.
But by that time the damage had been done. Instead of inserting Parnell at the beginning of the inning, when the Leverage Index was 2.26, Collins opted to stick with the inferior Rice because there was no save situation; and heaven forbid you use your closer in a non-save situation.
As a result, Nick Punto doubled, Adrian Gonzalez advanced him to third on a groundout, Matt Kemp was intentionally walked, Andre Ethier singled Punto in, Parnell came in, Juan Uribe drove home Kemp with a base hit, and the inning was ended when Ramon Hernandez grounded into a double play.
The Dodgers, on the other hand, used their best reliever, Kenley Jansen in the eighth inning, the highest leverage situation (1.79 LI heading into the inning, but it escalated to 3.65 by the inning’s end). Jansen was able to retire the Mets without allowing a run, and ended up getting the win when the Dodgers rallied off Rice in the ninth. Don Mattingly brought in the second-best available reliever, Proven Closertm Brandon League, to get the save in the bottom of the ninth; a slightly lower leverage situation (1.71 LI).
One could make the argument that even if Parnell had started the ninth, there’s no guarantee that he wouldn’t have allowed the Dodgers to score as well, but one thing is undeniable: Parnell gives the Mets a better chance to win the game than Rice and his 15.1% walk rate and 88 mph fastball does.
The Dodgers won because they used their best reliever when the most was at stake, and the Mets didn’t.
How many games does the Mets bullpen have to blow before something new is tried? How many times do we have to watch the 7th best reliever on the team surrender a lead in the seventh or eighth because the closer has to be used in a save situation? Why does Scott Rice have 13.1 innings pitched while the much better Bobby Parnell only has 9.1? How long does this madness have to go on before it’s over?
Only time will tell. Until then, here’s to Greg Burke coming in with two on in a tie game in the seventh.
Joe Vasile is the host of “Ball Four with Joe Vasile” on 91.3 FM WTSR in Trenton, airing Tuesdays from 12-1 p.m. Follow him on Twitter at @JoeVasilePBP.
BRAVO!
Thanks Chris.
Although, I seem to recall when someone looked into this last year (Brian?), as LI goes up, so does Parnell’s ERA. In any event, I agree with you that the set ways of running the relief corps could use some real improvement. In our case firing Warthen and TC would be a big plus alone.
Where did you see this? I did a search and I didn’t see anything (on this site, anyway) that came to that conclusion. If you can dig it up I would be interested in reading that.
It was probably me since I am Parnell’s biggest doubter. The stats for LI are easily found in baseball reference under splits, and you can see they have not been pretty in year’s past, including last year.
Ah, okay.
Well, that’s kind of a tautology. When LI goes up, it’s generally because you’ve got men on base/men in scoring position. It’s pretty obvious that if guys are already on base that the pitcher’s ERA would go up. That’s probably true of every pitcher in baseball with a large enough sample size.
Does this solutioj mean we use Parnell evey day and blow out his arm???
Read the footnoted passage from Keri’s book and you’ll have my answer to that.
So you’re implying that they are keeping Jansen in the setup role so that they can use him anytime instead of only in save chances… it would be very intersting if that indeed is the main reason for them doing it.
Could it also be a ploy to save money as well? Closers get paid ridiculous money, and by keeping him out of the closer’s role, that would keep his price tag lower than it would be if he were a closer. But then there is the ethical dilmena of screwing around with a player’s legacy and career earnings to save a few bucks.
Until saves are not the measure of sucess and measuranment of payday for relievers, this will never come nito fruition.
The money is certainly a big (if not the driving) factor in maintaining the status quo at this point. Not only are the pitchers hesitant to do this in fear of leaving money on the table, but managers as well. If they try something this different, and it doesn’t work out for them, they are out of a job.
The whole “closer” thing can be traced to Tony La Russa with Dennis Eckersley. He can be thanked for this “pigeon whole”. I agree completely that the team’s best reliever should be used in situations when the game is on the line, rather than when there’s a three-run lead or less in the ninth. Scott Boras and company would do everything in their power to prevent it, due to the money factor, but I’d love to see what one of our saber-geeks like DePodesta or Riccardi could come up with.
That is the thing. Relief pitchers would be hesitant to change to a system like this, because Closers make so much money. By sacrificing some saves this season, they are potentially leaving millions on the table in free agency. This has to be experimented with with a bullpen like ours, where there is nobody who you look at and say “definite closer”, just one guy who is really good and others who are okay. Obviously with success in this system, you’d be able to better lure free agent relievers to the team to bolster the ‘pen even further.
[…] by metman726 [link] [comment] Source: Reddit No Comments. « WFAN […]
So what happens after Parnell gets out of the inning and the Mets don’t score? The bull pen gives up the run the following inning prolonging the agony. Better scenario was Niese pitching against the Phillies with Howard on the bench. Which would you rather see? A tiring Niese on the mound or a righty coming in to pinch hit late in the game ?(knowing that Manuel would use a red hot Howard off the bench). It isn’t so much the bull pen(we know it’s bad) but Collins’ decision making doesn’t help matters either.
I agree: the Collins’ decision making when it comes to bullpen management has been atrocious, but the overall idea is to not have Scott Rice on the mound with the game on the line. You want your best available pitcher on the mound in the biggest spots.
Joe, He did that. And as expected, with a tight game on the line, boom, out goes a double. Sure Cowgill botched the read, but Parnell looked sad with a big save on the line, and several innings of spotless relief. When its “game on”, Parnell is not the guy you want out there IMO. That said, theres no one in the pen as presently configured that I would.
Um…Parnell got an easy fly ball to CF. Our CF couldn’t catch it. And you blame the pitcher?
So Joe! What do you say about tonight’s game? Bull pen blows the game twice! Pretty cool,huh?!
I’d say the bullpen was horribly mismanaged by Collins, as usual. Using two pitchers for a third of an inning each in the sixth? That was what necessitated Marcum coming in in the 14th. As for the blown save in the ninth, that is as much, if not more, on Cowgill and Baxter for the atrocious outfield defense than it is on Parnell.
.324 — Juan Pierre v. Mets
.368 — Justin Ruggiano v. Mets
Even when we face the Marlins we can count on these guys to inflict pain predictably.
Small Sample Size, especially when it comes to Ruggiano.
If the Mets continue with these type of abysmal games I don’t see how the team can hold on to Collins until the end of the year. They’ll have to make an assessment at the All-Star break and readjust their thinking. Collins is managing like a desperate man. No logic, no consistent game plan. Ike should be sent down to the minors. Not to punish him but to remind him that Duda can fill in just as well if not better. Light a fire under him and stop with the excuses. His approach at the plate is all screwed up. Why wait until Ike gets going? By the time he does the team will be 10 games under.
Unfair/inaccurate analysis. The bullpen ERA sucks, but it’s mainly because of a bad trip in Colorado combined with terrible outfielders. If our OFs don’t get to the ball, the pitchers get hurt. Duda, Baxter, and Cowgill have been the biggest culprits so far. By sending Ike down, moving Duda to first, and putting out an OF of Valdespín in LF/CF, Lagares in CF, and Cowgill/Baxter platoon in RF, we would dramatically improve our defense while also gaining a bit on offense. Really, though, wouldn’t you expect a Valdespín+Lagares+Cowgill/Baxter defense to be better than Duda+Cowgill+Baxter/Byrd?
FWIW – I’ve been pleasantly surprised by Byrd’s outfield defense.
[…] and more strictly, the Mets might be a marginally better team, especially when combined with the other changes I’ve […]