Closers to the Edge

Around the time of Dennis Eckersley‘s conversion from starter to closer, relief pitching underwent a reinvention. Gone were the Gossage/Fingers types, so-called “relief aces,” who were counted upon to shut down the opposing team at a crucial time, typically for more than merely three outs. In their place came closing specialists, high-octane guys that entered the game, (hopefully) recorded their three outs and called it a night. This has been commonplace for more than two decades.

Nowadays, fans wait with bated breath when their team’s 9th-inning guy ventures on past his typical assignment. For good reason too.

In 2013, there have been 36 instances where a closer has pitched past recording his third out, accounting for a total of 20 innings. In these 20 innings, the combined WHIP of these pitchers has been 2.00 and their ERA sits at a brutal 6.75.

The most notable offender of the bunch is Fernando Rodney, the Tampa Bay Rays closer that has seemingly returned to his pre-2012 ways (although he has been dominant in his last 6 appearances). His four extended pitching appearances (from here forth referred to as EPs) were good for five earned runs in 2.2 innings.

Los Angeles’s Ernesto Frieri has led the way with six(!) EPs, totaling the same 2.2 innings, but only allowing one earned run and striking out four.

Only 16 of the game’s 30 closers have been tapped for a fourth (or greater) out in a save situation, and often it appears that was indeed a foolish choice. However, strikeout rate is pretty high for EPs, with 26 Ks coming in the 20 innings, good for an 11.7 K/9.

Below is a table of the EP leaders:

Rk Player Tm EP IP H R ER BB SO ERA
1 Ernesto Frieri LAA 6 2.2 2 1 1 2 4 3.38
2 Bobby Parnell NYM 5 2 3 3 2 2 1 9.00
3 Fernando Rodney TBR 4 2.2 4 5 5 4 3 16.88
4 Tom Wilhelmsen SEA 3 2.2 2 0 0 1 5 0.00
4 Edward Mujica STL 3 1.2 2 1 1 0 2 5.40
4 Kevin Gregg CHC 3 1 1 0 0 1 1 0.00
7 Sergio Romo SFG 2 0.2 0 0 0 0 0 0.00
7 Steve Cishek MIA 2 1.1 3 4 4 3 0 27.00
9 Addison Reed CHW 1 2 0 0 0 1 4 0.00
9 Grant Balfour OAK 1 1 1 0 0 1 1 0.00
9 Jonathan Papelbon PHI 1 0.2 2 0 0 0 2 0.00
9 Heath Bell ARI 1 0.2 0 0 0 0 1 0.00
9 Joe Nathan TEX 1 0.1 0 0 0 0 0 0.00
9 Jose Veras HOU 1 0.1 0 0 0 0 1 0.00
9 Brandon League LAD 1 0.1 1 1 1 0 1 27.00
9 Jim Johnson BAL 1 0 2 1 1 2 0 INF
Provided by Baseball-Reference.com: View Play Index Tool Used
Generated 7/1/2013.

Addison Reed‘s lone EP is a fascinating one, coming on June 5th in a 16-inning mess against the Seattle Mariners. Despite doing fine work in his final two innings, Reed was tagged with 5 earned runs in the 14th frame and still allowed to continue (and pick up the almighty win!), as he was the 8th pitcher used by the White Sox.

For comparison’s sake, in Goose Gossage’s brilliant 1975 campaign, 43 of his 62 appearances went beyond three outs.

Granted, 20 innings is a small sample to draw on, especially when culled from 16 players. But as far as the first half of the 2013 season is concerned, closers are performing poorly as a whole when asked to go beyond their expected duty.

Note: For my purposes, closer is defined as team’s current saves leader.

Leave a Reply

21 Comments on "Closers to the Edge"

Notify of
avatar
Sort by:   newest | oldest | most voted
Bryan O'Connor
Editor

Dalton, I’d be interested to see how many of these “EPs” were closers pitching into the 10th (or later) after completing their typical ninth inning (and probably blowing a save) and how many were guys brought in with an out or two in the 8th and runners on base.

It seems to me (with no data to back up this claim) that managers are more willing to let a closer pitch a second inning after a quick ninth than to use him in a high-leverage situation in the 8th, where Gossage and the “firemen” of his ilk earned that moniker.

e pluribus munu
Guest

Looking over the B-R game logs of the first six pitchers on the list (24 EPs, or 2/3 of Dalton’s total – one EP for Mujica discounted for not occurring in the closer role), fifteen involved closers entering in the 8th, while in the remaining nine cases the closer entered in the 9th or later.

Brent
Guest

In Dan Quisenberry’s 1983 season, he had 69 appearances for 139 innings, 45 saves, a 1.94 ERA, a .928 WHIP and 49 of the 69 appearances were for more than 3 outs, including a 16 out outing against the Yankees on July 22, a game which he ironically did not finish (he got the Win after pitching from 2 outs in the 6th through the 11th and Steve Renko got the save for getting the last 3 outs in the bottom of the 12th)

Timmy Pea
Guest

Wow, that is an incredible year. I had to look it up. You know who I think never gets enough credit for being a good pitcher is Kent Tekulve.

Timmy Pea
Guest

Nothing against Bruce Sutter, but Quiz was at least as good if not better than Sutter and Sutter is in the HoF. Quiz doesn’t even have his number retired by the Royals.

John Autin
Editor

I second both of Timmy’s points — Sutter has nothing over Quiz, and Teke is sadly forgotten.

My favorite Tekulve stat: 90 games, 105 innings, at age 40. There have been 9 seasons of 90+ games; Tekulve and Marshall own 3 each.

Ed
Guest

Small typo in the second sentence: “crucial team” instead of “crucial time”.

Ed
Guest

Indians closer Chris Perez has pitched exactly one inning in 18 of his 19 appearances this year. The lone exception was on May 26th against Boston when he entered the ninth with 5-2 lead and was pulled after blowing the lead and only recording two outs.

The lack of imagination displayed by baseball managers can be quite frustrating.

John Autin
Editor

Uh-oh, someone pressed my “John Hiller” button….

Darien
Guest

A least it wasn’t your “John Rocker” switch.

bstar
Guest

Dalton, great job at getting these facts out there. They fly in the face of the idea that modern 1-inning guys can just magically start pitching 2 or 3 innings on command. It doesn’t work that way.

They would have to learn how to pitch differently and dial it down a tad. And that doesn’t happen overnight.

e pluribus munu
Guest
bstar, I’m not sure these data alone can get us far towards that conclusion. For example, looking just at Frieri, his so-so ERA in EPs involves three runs. One comes from the initial batter he faced in the 8th – a home run; in the ninth, he struck out all three batters. (The other two come from 9th inning walks; one of those came around via advancing on a fly, an SB, and a single – followed by an out for the save.) Five of Frieri’s six EPs were saves, and the BS allowed one run for the tie. Frieri… Read more »
bstar
Guest
You’re probably right about not enough evidence, although re-reading my comment I don’t think I specifically implied it was “extra work” that was the cause. I think getting pitchers out of their comfort zone and out of their routine are more likely causes than extra effort in the second inning of work. But there’s some games out there that are not showing up in Dalton’s search. Those are the appearances by closers brought into the game in the eighth with the expectations of said closer finishing the game. Instead, he gets touched up in the eighth, blows the save, and… Read more »
bstar
Guest

Let me give a couple more examples of pitchers being asked to pitch out of their comfort zone and not having as much success with it:

-asking starting pitchers to pitch on 3 days rest in the postseason

-asking starting pitchers to pitch in relief in the postseason

e pluribus munu
Guest
You make a good point, bstar. Frieri’s record had no pre-9th inning blown saves to draw my attention to the issue. As for the larger question, you’re certainly correct that, as in most skill performance, the likelihood of player failure rises when the context is unfamiliar. But is the best approach to stick with the current high level of specialization? Rosters are limited and situational variables are not, so it figures that there’s a point past which narrow role definition becomes counter-productive. As you say, these guys aren’t robots, but the more you suppress the demand for adaptability, the more… Read more »
bstar
Guest
JA correctly points out another source of probable bias in Dalton’s numbers @17, so there’s that. We talked about this whole thing a while back, and someone brought up an article by Rob Neyer in which he talked about the modern bullpen. The sentence that really stuck with me was, (paraphrasing), ‘Modern bullpens are probably not maximized from an efficiency standpoint, but they almost certainly are maximized from a practicality standpoint.’ Brilliant. I’m all for trading 3-run-lead saves for the occasional 4- or 5-out save. But the seventh inning? You can’t swap those out one-for-one because there are waaay more… Read more »
Ed
Guest
Bstar @19 – Two points: 1) You may be aware of this already but Jos Pos did an analysis earlier this year looking at win rates in the 9th inning going back to the 1950s. He basically found that the win rate is the same every decade, even though bullpen usage has changed over time. Something to think about: http://joeposnanski.blogspot.com/2013/03/the-closer-you-get.html 2) I think my problem with the 9th inning closer model is this…all managers act as if they have a 9th inning shut down guy even when they don’t. If you have Mariano Rivera or Craig Kimbrel, or someone else,… Read more »
e pluribus munu
Guest
Posnanski’s data showing constant rates for ninth-inning one-run lead win rates is interesting, very counter-intuitive, and really nicely researched. It bears directly on your second point, Ed, in the sense that we know there have been some very effective closers in recent decades, and if the average win conversion rate is constant despite their success, then uniformity in practice across teams of varied closer quality is costing teams that don’t have the right personnel to a degree greater than the practices in previous decades. As for Neyer’s comment, bstar, I think it presumes the uniform validity of the current model… Read more »
Doug
Editor
In addition to Addison Reed, Paul Clemens of the Astros recorded a relief win this season when allowing 5+ ER. Clemens’ victory also came against the Mariners; he pitched innings 5 through 8 on April 9th, entering with a 13-0 lead and leaving with a 15-5 cushion. Prior to Clemens and Reed, the last relief win when allowing 5+ ER was by the Yankees’ Ed Whitson in 1986. Jim Slaton of the Brewers had two such wins in the same month in July 1983. Carl Erskine did this twice in 8 days for the 1949 Dodgers; between those two games… Read more »
John Autin
Editor

Dalton, this is interesting. Did you check for selection bias? I believe the situation in which a modern closer is most likely to go beyond 3 outs is when he’s already blown the save, and the game is tied in extras.

wpDiscuz