We’ve talked more than usual about the bullpen this year, in part because it’s been so much better than expected or feared. We still don’t have a great way to measure relievers, though. Typically, we want to have one number that we can point to and say – see! But, in my opinion, WAR doesn’t do the job here as good as it does for position players. My go-to stats for relievers are usually ERA, WHIP and OPS allowed. Here’s a chart for all of the Mets relievers who’ve amassed 50 IP of relief since 2010:

Results
Rk Player Year GS WHIP IP G SV ERA FIP ERA+ BAbip OPS OPS+
1 Edwin Diaz 2019 0 1.379 58.0 66 26 5.59 4.51 74 .381 .834 125
2 Jeurys Familia 2019 0 1.733 60.0 66 0 5.70 4.88 73 .350 .831 129
3 Robert Gsellman 2019 0 1.366 63.2 52 1 4.66 4.12 89 .317 .766 109
4 Seth Lugo 2019 0 0.900 80.0 61 6 2.70 2.70 154 .267 .562 53
5 Robert Gsellman 2018 0 1.300 80.0 68 13 4.28 3.95 87 .294 .700 101
6 Paul Sewald 2018 0 1.509 56.1 46 2 6.07 4.23 62 .335 .820 133
7 Seth Lugo 2018 5 1.076 101.1 54 3 2.66 3.17 140 .270 .595 71
8 Hansel Robles 2017 0 1.341 56.2 46 0 4.92 5.13 84 .264 .750 105
9 Paul Sewald 2017 0 1.209 65.1 57 0 4.55 3.74 91 .292 .706 92
10 Josh Smoker 2017 0 1.704 56.1 54 0 5.11 4.92 81 .355 .858 135
11 Addison Reed 2016 0 0.940 77.2 80 1 1.97 1.97 204 .290 .536 48
12 Hansel Robles 2016 0 1.352 77.2 68 1 3.48 3.56 116 .308 .703 95
13 Jeurys Familia 2016 0 1.210 77.2 78 51 2.55 2.39 158 .307 .574 62
14 Hansel Robles 2015 0 1.019 54.0 57 0 3.67 3.91 104 .228 .659 85
15 Sean Gilmartin 2015 1 1.186 57.1 50 0 2.67 2.75 143 .306 .626 79
16 Carlos Torres 2015 0 1.370 57.2 59 0 4.68 3.53 82 .327 .743 111
17 Jeurys Familia 2015 0 1.000 78.0 76 43 1.85 2.74 207 .273 .569 62
18 Carlos Torres 2014 1 1.309 97.0 73 2 3.06 3.86 113 .305 .715 111
19 Jeurys Familia 2014 0 1.177 77.1 76 5 2.21 3.07 157 .269 .587 74
20 Bobby Parnell 2013 0 1.000 50.0 49 22 2.16 2.33 166 .268 .555 62
21 Scott Rice 2013 0 1.353 51.0 73 0 3.71 3.40 97 .295 .634 87
22 LaTroy Hawkins 2013 0 1.146 70.2 72 13 2.93 3.06 122 .305 .656 90
23 Jon Rauch 2012 0 0.988 57.2 73 4 3.59 3.89 106 .224 .617 74
24 Ramon Ramirez 2012 0 1.461 63.2 58 1 4.24 3.93 90 .297 .698 99
25 Bobby Parnell 2012 0 1.238 68.2 74 7 2.49 2.99 153 .308 .648 84
Rk Player Year GS WHIP IP G SV ERA FIP ERA+ BAbip OPS OPS+
26 Pedro Beato 2011 0 1.284 67.0 60 0 4.30 4.22 86 .262 .719 105
27 Bobby Parnell 2011 0 1.466 59.1 60 6 3.64 3.21 101 .339 .679 94
28 Francisco Rodriguez 2010 0 1.151 57.1 53 25 2.20 2.63 179 .296 .597 68
29 Pedro Feliciano 2010 0 1.532 62.2 92 0 3.30 3.22 119 .351 .702 99
30 Raúl Valdés 2010 1 1.466 58.2 38 1 4.91 4.31 80 .313 .791 120
Provided by Stathead.com: View Stathead Tool Used
Generated 5/13/2021.

When choosing the filters, GS less than or equal to five was used. Now, it’s possible that a reliever made more than five starts and still amassed 50 IP in relief. But it gets us 2018 Lugo, which was the guy who was in the front of my mind.

All of those show something; yet, somehow, we are still left wanting more.

One thing that adds to the discussion is the number of times that a reliever allows a run in an appearance. Because relievers don’t pitch a lot of innings, a couple of bad outings can really destroy their ERA. Or, you can have a guy with a half-decent ERA but who never gives up multiple runs in an inning. Is it better to give up 5 R in one outing or 5 R in five outings?

Anyway, here’s the 30 pitchers above and the percentage of their appearances where they gave up runs. The dashed lines indicate a change in year. The first four are from 2019 and after the dashes is 2018.

Diaz – 29%
Familia – 35%
Gsellman – 44%
Lugo – 21%
—–
Gsellman – 57%
Sewald – 49%
Lugo – 31%
—–
Robles – 33%
Sewald – 33%
Smoker – 33%
—–
Reed – 16%
Robles – 29%
Familia – 21%
—–
Robles – 32%
Gilmartin – 20%
Torres – 27%
Familia – 14%
—–
Torres – 31%
Familia – 22%
—–
Parnell – 18%
Rice – 21%
Hawkins – 24%
—–
Rauch – 25%
Ramirez – 31%
Parnell – 27%
—–
Beato – 40%
Parnell – 32%
—–
Rodriguez – 17%
Feliciano – 21%
Valdes – 35%

A couple of things jump out to me from seeing this list. First, it breaks down at the edges, for the guys who consistently pitch less than or more than an inning. It’s probably best to ignore this for LOOGYs and long men. The second is that you want your closer to be around 20%. Rodriguez, Parnell and 2015 Familia were all below that line. 2014 and 2016 Familia was just above it and 2019 Diaz was way above at 29%. Finally, if you’re putting up a percentage in the 40s as a standard 21st Century one-inning reliever, it’s probably time to go.

Let’s check the percentages on the 2021 relievers with at least five appearances:

Diaz – 21%
Trevor May – 21%
Miguel Castro – 23%
Aaron Loup – 18%
Familia – 8%
Jacob Barnes – 33%
Gsellman – 38%

A few days ago, this would have looked better for Barnes. In his last outing on May 9, he allowed a run when he was sent back out for a second inning. If these percentages were calculated on May 8, he would have checked in with a 25% mark, compared to a 5.00 ERA.

10 comments on “Mets relievers and runs allowed per appearance”

  • BoomBoom

    Gsellman has also been pretty locked in since his first or second appearance. The eye test says this is the best he has looked in years.

  • Metsense

    The primary job of a relief pitcher is to not allow runs to score when they appear in a game. Tom Tango is “The Book” stated an average relief pitcher give up runs 30% of the time and an elite reliever gives up runs 21% for a one inning appearance. I couldn’t find this statistic anywhere. Thanks for doing work Brian. As you can see there are five “elite” pitchers in the 2021 bullpen.
    ERA is pretty much useless for a reliever. WHIP, HR/9 and K/9 are helpful in judging a reliever but % run/appearance is what a reliever is paid for.

    • Brian Joura

      Thanks for the info Metsense!

      It’s important to note that none of the numbers presented in the article differentiated between innings that the reliever threw. If he threw 3 innings and allowed 1 run – it counted as an appearance where he gave up a run. And if a player went 0 or 0.1 IP and did not give up a run – it counted, too. So, we can’t exactly compare to the number you quoted from Tango. But it’s definitely good to have that number.

  • JamesTOB

    Fascinating, thanks Brian.

  • Wobbit

    If Diaz can close 4 out of 5 chances, I’m happy.

    • Brian Joura

      In his big year with SEA, Diaz had a 93.4 save percentage. And that came after an 87.2 save percentage the year before. Maybe I’m greedy but I can’t be happy with an 80% save percentage from him.

      • BobP

        And he was at 79% in 2019. I don’t want to live through that again.

  • MattyMets

    This is great. ERA is particularly useless for relievers early in the season or in a short-season….WHIP is good, but it wouldn’t have made John Franco look good as he was known for putting a runner on so he could get a DP, or putting 2 or 3 on before escaping unscathed (most of the time)….I forgot how bad Josh Smoker was. It’s easy to fall in love with a hard-throwing lefty, but boy was he hittable. I think he might still be out there….A stat I’d like to see is percentage of inherited runners allowed to score. This, obviously is not reflected in ERA.

  • Wobbit

    80% closer rate is good for me, especially since not every one of the “blown saves” becomes a loss.

Leave a Reply

Your email address will not be published. Required fields are marked *

The maximum upload file size: 100 MB. You can upload: image, audio, video, document, spreadsheet, interactive, text, archive, code, other. Links to YouTube, Facebook, Twitter and other services inserted in the comment text will be automatically embedded. Drop file here