Introducing The Reliever Reliability Quotient (RRQ), A New Stat To Help Determine Closers At Risk & More

by Eric Stashin (aka The Rotoprofessor)

For the first time ever I am debuting my own statistic that we will utilize moving forward (I’ll be posting the latest stats every Monday morning).  It is a work in process, so I’ll be tweaking the formula a bit as we move forward, but the purpose is to rank relievers and their reliability to produce, whether it’s as a closer or just in general, based on their underlying skills and not their actual results.  Let’s call it the “Reliever Reliability Quotient”, which I’ll refer to as RRQ moving forward.  I’d love to hear your feedback of our first run, and your thoughts in general, so please share them in the comments!

Among the stats that are taken into account in calculating the number are:

  • Line Drive Rate
  • Groundball Rate
  • Fly Ball Rate
  • Popup Rate
  • Strikeout Rate
  • Walk Rate
  • O-Swing%
  • SwStr%

Each skill is weighted differently, for how important I view it towards the makeup of an elite reliever.  Just because you walk a few batters or are currently carrying an elevated line drive rate doesn’t mean you can’t be successful, so we have to adjust accordingly.

The baseline is a score of “0”, with the potential for pitchers to be both positive and negative.  Even with a high score (or a low score) there is no certainty in a player’s production.  The stat is meant as a way to try and help us identify those who are truly at risk of losing their job, to identify those who we’d expect to rebound (assuming they continue to get an opportunity to produce) or to find the most likely to succeed if given a chance. 

Let’s take a look at how I would look at the scores:

Score
Skill Level
76+Elite
51-75Above Average
26-50Solid (though worth monitoring)
0-25Questionable makeup
<0High Risk For Struggles

Now a few notes on who’s included in this first run (which is just the current MLB closers):

  • Any reliever with at least 20 IP were included
  • Aroldis Chapman doesn’t have enough innings pitched, and was therefore left off the rankings
  • Given the state of the bullpens in Washington and San Francisco, neither have a representative
  • I included both Andrew Miller & Cody Allen
  • I included Bud Norris, who should be the answer in Los Angeles once he’s healthy

With that in mind, let’s take a look at the rankings of the current closers:

Pitcher
RRQ
Craig Kimbrel98.5
Felipe Rivero83.37
Roberto Osuna79.6
Kenley Jansen76.97
Wade Davis65.32
Jim Johnson52.7
Corey Knebel47.92
Andrew Miller47.4
Edwin Diaz45.87
Bud Norris45.2
Ken Giles41.42
David Robertson24.27
Brad Brach15.8
Justin Wilson13.17
Greg Holland11.82
Brandon Maurer5.2
Brandon Kintzler2.85
Raisel Iglesias2.65
Mark Melancon2.55
Matt Bush0.72
Kelvin Herrera0.45
Hector Neris-3.93
Fernando Rodney-7.13
Addison Reed-8.7
Alex Colome-13.45
A.J. Ramos-13.6
Santiago Casilla-24.6
Seung Hwan Oh-38.55

Thoughts:

  • Iglesias ranked #18? While he looks like he has elite stuff, he has had some control issues.  He also got hurt by a pedestrian O-Swing% (30.90%, where the average for a reliever was 30.77%) and a lack of popups (3.30%, compared to the average of 9.98%).  It’s the popups that have the biggest impact on dragging him down the rankings (and his mark is actually the third lowest of any pitcher included on the above chart).
  • Surprised to see Osuna ranked among the elite in the game? His score is the second highest, even when we look at setup men, and it’s hard to argue.  He doesn’t walk people, he has been fooling them at a high level (48.20% O-Swing%) and he strikes out a ton of batters (37.00% K%).  Almost everything about him screams elite.
  • Then you move down one line and see Felipe Rivero?! He’s getting a lot of swinging strikes (15.50%) and avoiding walks, but that’s not all.  He’s a groundball (58.3%) and popup (13.3%) machine, and that means there’s little risk of home runs.  Maybe the numbers regress, but with the way he’s throwing right now he’s clearly elite.
  • The other big surprise is likely Colome being ranked so low. Basically, in the stats that we looked at to form our score he’s relatively average, or below average, across the board and we’ve seen things start to catch up to him of late.  That’s not to say that he won’t rebound, but there’s risk that he continues to struggle.  While we wouldn’t push the panic button, it’s definitely something to monitor.

Sources – Fangraphs, MLB.com

Make sure to check out all of our updated rankings:

Position
Posted
CatcherJune 1
First BaseJune 3
Second BaseJune 5
Third Base--
Shortstop--
Outfield--
Starting Pitcher--
Relief Pitcher--

7 comments

  1. Bill says:

    Very interesting. This generally matches my sense of the relative pecking order, although there are some big surprises (expected Holland to be 6-8 spots higher, didn’t think Johnson would be nearly so high).

    Also, very interesting that only about 10 closers rate as solid or better. I don’t disagree, but I also don’t have the sense that it’s a view that’s widely shared.

    • Rotoprofessor says:

      I would agree that most people view more as “solid”, and I’m not necessarily going to disagree, though the underlying skills do give me at least some cause of concern. Prior to this recent bump most would’ve viewed Colome among the elite, but I’m guessing that sense has changed over the past few weeks.

      We’re going to be using this stat to try and sort through all the “questionable” situations, and also try to identify potential candidates to emerge. It should bring some interesting discussion (we’ll be using it to look at the Rangers’ closer candidates a little bit later this afternoon, for example)

  2. Gern says:

    Hey Prof, interesting work. Have you regressed these variable on ERA and WHIP? My guessis your overall model is theses variables predict level II success (ERA & WHIP) which predicts level III success (saves or holds).

    • Rotoprofessor says:

      It’s not really meant as an ERA/WHIP indicator, though I could see how it would apply. Instead it’s more analyzing the skills that are needed to be successful and if they are present (which would hopefully lead towards SV/HLD).

      • Gern says:

        Thanks Prof, I was just curious about the model. I would guess adding ERA and WHIP (weighted as you see fit might provide even more power to better explain “success” (i.e, saves or holds)…

  3. Michael BASH says:

    I don’t agree on the ranking of Holland, “Questionable makeup” I think he should be in the “solid” range at least, he’s leading the league in saves and was a elite reliever pre-TJS

    • Rotoprofessor says:

      It’s fair, but remember pitchers with “sub-par” skills can still be successful. Sooner or later, though, things could catch up to them.

      Where Holland is hurt is mostly the inflated fly ball rate, and that could come into play significantly in Coors Field. There’s the risk of his HR allowed rising, and as he’s also had some BB issues there could be reason for concern at some point.

Leave a Reply

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