How to Target Middle Relievers

Note: Make sure to read Piotr’s comment. It’s a simpler in-season alternative.

A completely underappreciated class of players is non-closing relievers. With starters going fewer and fewer innings, the gap between the starting dregs and good non-closers has narrow considerably. This past season Devin Williams, Matt Foster, Josh Fleming, and Pete Fairbanks each ended the season as a top-60 pitcher value.  Starters don’t throw enough to get the Win or accumulate many strikeouts. A reliever can easily occur more strikeouts than a single-start starter in a single week. Additionally, the strikeouts usually come with better ratios and the off chance for a Win or a Save. Picking out elite middle relievers is tough, but I have a couple of simple rules to follow to help find them.

Determining who is going to be a good middle reliever in the upcoming season is tough. It’s almost impossible. First, few repeat being lights out each season (e.g. Andrew Miller). Second, the relievers who show the skills to be elite from season-to-season become closers and their price skyrockets. It’s a dynamic group.

The 2020 season added to the chaos. When it ended, MLB and fantasy teams were just settling into their rotations. Middle relievers were just getting to the point of becoming roster regulars. I don’t think they were as valuable in the shorter season where teams had excessive FAAB to spend on a new pitching staff each week.

The question remains, how many middle relievers should a fantasy manager focus on. The key is the league and roster size. In shallow leagues, not even all the closer might be rostered, so middle relievers are not even a thought. In Only leagues, nearly every pitcher is rostered, so managers will be scrambling for anyone who shows any sign of life. I’m going to focus on 12 to 15-team leagues with the standard NFBC 23-man roster and 7 man bench… the middle ground.

In Saves leagues, fantasy managers are going to want two closers with a rotating additional Saves speculative add (or handcuff). While this speculative add could be streamable, he might be a Saves-only play (high ratios, few strikeouts). I’d go one step further and get one more elite reliever. If several are similar, roster the one who might get a Save or two (i.e. Tampa bullpen arms). The goal with this pitcher is to provide better strikeouts and ratios than 6th or 7th starters.

The problem is that it’s tough to identify such pitchers before the season starts. This point became clear when I was re-examining my draft-and-hold teams. They can be useful when good. After going through several iterations trying to find a decent stat target, I settled on projected ERA under 3.25 and a WHIP under 1.25. Here is a list of non-relievers coming into 2020 ranked by projected Saves.

Top Relievers Based on Talent
Proj Actual
Name IP ERA WHIP W SV IP ERA WHIP W SV
Kirby Yates 23.0 3.40 1.13 1 12 4.1 12.46 2.54 0 2
Liam Hendriks 26.0 3.14 1.07 2 11 25.1 1.78 0.67 3 14
Roberto Osuna 21.0 3.61 1.12 1 11 4.1 2.08 0.69 0 1
Aroldis Chapman 20.0 3.04 1.13 1 11 11.2 3.09 0.86 1 3
Ken Giles 24.0 3.53 1.16 1 11 3.2 9.82 2.18 0 1
Taylor Rogers 24.0 3.50 1.18 1 11 20.0 4.05 1.50 2 9
Edwin Diaz 24.0 3.04 1.07 1 11 25.2 1.75 1.25 2 6
Jose Leclerc 24.0 3.52 1.23 1 10 2.0 4.50 2.00 0 1
Josh Hader 28.0 3.10 1.07 2 9 19.0 3.79 0.95 1 13
Nick Anderson 24.0 3.26 1.09 1 8 16.1 0.55 0.49 2 6
Will Smith 21.0 3.50 1.18 1 4 16.0 4.50 0.94 2 0
Austin Adams 19.0 3.19 1.20 1 3 4.0 4.50 1.25 0 0
Jose Alvarado 22.0 3.31 1.30 1 3 9.0 6.00 1.67 0 0
Ryan Pressly 24.0 3.35 1.15 1 2 21.0 3.43 1.33 1 12
Ryne Stanek 24.0 3.74 1.25 1 2 10.0 7.20 1.90 0 0
Matt Barnes 25.0 3.39 1.23 1 2 23.0 4.30 1.39 1 9
Corey Knebel 19.0 3.72 1.26 1 2 13.1 6.08 1.73 0 0
Diego Castillo 22.0 3.25 1.18 1 2 21.2 1.66 1.06 3 4
Dellin Betances 20.0 3.22 1.18 1 1 11.2 7.71 2.06 0 0
Will Harris 23.0 3.67 1.26 1 1 17.2 3.06 1.70 0 1
Blake Treinen 23.0 3.61 1.28 1 1 25.2 3.86 1.21 3 1
Seth Lugo 24.0 3.74 1.19 1 1 36.2 5.15 1.36 3 3
Ty Buttrey 24.0 3.51 1.20 1 1 26.1 5.81 1.41 2 5
Pete Fairbanks 19.0 3.58 1.24 1 1 26.2 2.70 1.39 6 0
James Karinchak 21.0 3.59 1.29 1 1 27.0 2.67 1.11 1 1
Chris Martin 19.0 3.47 1.14 1 0 18.0 1.00 0.61 1 1
Jake Diekman 23.0 3.51 1.30 1 0 21.1 0.42 0.94 2 0
Drew Pomeranz 21.0 3.59 1.22 1 0 18.2 1.45 1.02 1 4
Joe Kelly 24.0 3.57 1.28 1 0 10.0 1.80 1.50 0 0
Oliver Drake 20.0 3.49 1.22 1 0 11.0 5.73 1.18 0 2
Andrew Kittredge 13.0 3.48 1.18 1 0 8.0 2.25 1.25 0 1
Luke Jackson 21.0 3.59 1.26 1 0 26.1 6.84 1.97 2 0
Tommy Kahnle 25.0 3.62 1.22 1 0 1.0 0.00 2.00 0 0
Emilio Pagan 24.0 3.70 1.13 1 0 22.0 4.50 1.05 0 2
Chad Green 22.0 3.49 1.08 1 0 25.2 3.51 0.82 3 1
Caleb Ferguson 14.0 3.62 1.25 1 0 18.2 2.89 1.02 2 0
Brusdar Graterol 17.0 3.62 1.25 1 0 23.1 3.09 0.90 1 0
Average 21.9 3.47 1.19 1.1 3.6 16.7 3.94 1.32 1.2 2.8

Now, there are going to be some injuries to deal with depending on when a manager drafts like Jose Castillo and Seranthony Dominguez, so risks (e.g. Dellin Betances) will need to managed. Some of the non-closers (e.g. Pressley and Lugo) ended by decent values by season’s end. But many didn’t live up to expectations. The results just get worse as the list is expanded. If I’m going to draft a middle reliever, it’s going to be from this list, but even then I’m going to be picky.

Just eyeballing the above list, about one third don’t come close to expectations, so how should a manager address the problem in-season. The answer is constant churning of the spot.

To find non-closers who excel, start with a criterion of being above average in K/9 and below league average in ERA and WHIP (link to sample). Additionally, I want pitchers used in high leverage situations so they’re likely to see a decent number of Saves+Holds+Wins. While the ratios and strikeouts are a fine base, hitting on a few Wins or Saves are ideal.

To locate such players, download the stats, open them in Excel, add a Filter to the header row (first image), and select Number Filter for the stats to above or below average (image 2).

Here are the results from the two lists taken from early in 2020.

Above-Average Relievers
Name (2 weeks in) IP K/9 ERA WHIP W+SV+HLD Name (4 weeks in) IP K/9 ERA WHIP W+SV+HLD
Joakim Soria 9.2 13.0 0.00 1.03 5 Liam Hendriks 16.1 12.7 1.10 0.67 12
Jake McGee 8.2 12.5 0.00 0.46 3 Brad Hand 10.2 11.0 3.38 0.94 11
Drew Pomeranz 7.2 11.7 0.00 0.52 8 Kenley Jansen 14.2 12.9 1.23 0.82 10
Tyler Duffey 8.0 12.4 0.00 0.25 6 Jordan Romano 14.2 12.9 1.23 0.89 9
Nick Anderson 7.1 13.5 0.00 0.68 6 Tyler Duffey 14.1 10.7 1.88 0.77 9
Gregory Soto 10.0 12.6 0.00 0.50 3 Jonathan Hernandez 18.1 11.3 1.96 0.87 9
Chad Green 11.0 12.3 0.82 0.36 6 Devin Williams 14.0 18.6 0.64 0.71 8
Trevor Rosenthal 10.0 11.7 0.90 0.90 5 Tanner Rainey 14.1 12.6 1.26 0.70 8
Jordan Romano 10.0 11.7 0.90 0.70 6 Chad Green 15.2 13.8 3.45 0.77 8
Kenley Jansen 9.1 11.6 0.96 0.64 6 Joakim Soria 15.2 9.8 1.72 1.15 7
Tanner Rainey 8.1 11.9 1.08 0.36 5 Scott Barlow 19.2 11.0 1.37 1.07 6
Caleb Ferguson 8.0 12.4 1.13 0.75 3 David Phelps 13.0 13.9 2.77 0.69 6
David Phelps 7.2 11.7 1.17 0.78 4 Caleb Ferguson 13.1 13.5 0.68 0.68 5
Liam Hendriks 11.1 13.5 1.59 0.88 8 Gregory Soto 17.1 10.9 3.12 0.87 5
Scott Barlow 13.0 11.8 2.08 1.08 4 A.J. Minter 12.1 11.0 0.73 0.89 4
Shawn Armstrong 8.2 10.4 2.08 0.81 4 Matt Foster 12.0 11.3 1.50 0.83 4
Jonathan Hernandez 12.2 10.7 2.13 0.87 7 Felix Pena 17.1 10.4 2.08 1.04 4
Nick Wittgren 8.0 11.3 2.25 0.88 3 Brent Suter 14.2 11.1 3.68 1.23 4
Burch Smith 12.0 9.8 2.25 0.67 3
Sergio Romo 7.0 11.6 2.57 0.43 7
Travis Lakins 9.1 10.6 2.89 1.07 4
Jonathan Loaisiga 6.0 12.0 3.00 1.17 3
Brent Suter 12.0 10.5 3.00 1.17 3
Nate Jones 7.2 11.7 3.52 1.17 3

In the first list, no one differentiated themselves from the pack, but within a month Williams (a top-30 pitcher end-of-season) and a couple of eventual closers (Romano and Soto) rose to the top. Just checking this list once a week can help an managers find a few cheap gems.

I’m not 100% sure if this is the best method. It’s better than my current method of hearing mentions of this or that reliever on a podcast or seeing one added my an opposing time. I really don’t want to spend too much time locating the best one. It seems futile with the position in such flux. In the end, I plan on going with just a couple of rules to find middle relievers.

  1. When drafting, only add the best of the best with an eye toward Saves.
  2. During the season, cheaply churn roster spots each week to find this year’s Devin Williams/Josh Hader/Andrew Miller.

I wish I had a better approach, but it’s close to the same approach to take with Starters and Closers. Draft the best ones and replace the subpar ones with better ones from the waiver wire.





Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.

5 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
mr_hoggmember
3 years ago

I also look for pitchers being elevated by elite teams, since those teams will generate more wins, saves, and holds, will have less garbage time, and will be less likely to leave a reliever on the mound when he doesn’t have his stuff. Of course, this path has led me in recent years to Carson Smith, Ryan Brasier, and other total flame outs … but hey it’s a plan …