The Most Improved Pitchers This Season by Projected WAR

What follows represents an attempt by the author to utilize the projections available at the site to identify the five starting pitchers whose per-inning WAR projections have most improved since the beginning of the season.

For every pitcher, what I’ve done is first to calculate his preseason (PRE) WAR projection prorated to 150 innings, averaging together Steamer and ZiPS forecasts where both are available. What I’ve done next is to calculate the prorated WAR for every pitcher’s rest-of-season (ROS) WAR projection (again, using both Steamer and ZiPS when available). I’ve then found the difference in prorated WAR between the preseason and rest-of-season projection.

Only those pitchers have been considered who (a) are currently on a major-league roster and (b) have recorded at least 50 innings at the major-league level and 20 innings this season and (c) are expected to work predominantly as a starter for the duration of the season. Note that PRE denotes a player’s combined Steamer and ZiPS preseason projection; ROS, the rest-of-season projection. Diff is difference between the prorated ROS projections and the PRE one. Data is current as of Tuesday.

5. Chase Anderson, RHP, Arizona (Profile)

Proj. IP K/9 BB/9 HR/9 FIP WAR
PRE 150 6.3 3.2 1.2 4.60 -0.6
ROS 150 7.2 3.1 1.2 4.45 0.6
Diff 0.9 -0.1 0.0 -0.15 1.2

Notes
Like some of the other pitchers who appear below, Anderson’s improved WAR projection isn’t merely a function of a more encouraging FIP projection, but also of a change in role. After struggling in a starting capacity at Triple-A over the first half of 2013, Anderson finished last year as a reliever — in which role he was partially projected for 2014. The same pitcher will typically record both a higher strikeout rate and lower ERA in a relief role than in a starting one. What Anderson has done, though, is simultaneously improve upon both projected figures while also transitioning back to a full-time starter’s role.

***

4. Josh Tomlin, RHP, Cleveland (Profile)

Proj. IP K/9 BB/9 HR/9 FIP WAR
PRE 150 5.4 1.7 1.3 4.33 0.0
ROS 150 6.6 1.6 1.3 4.16 1.2
Diff 1.2 -0.1 0.0 -0.18 1.2

Notes
Over his first four major-league seasons (about 350 innings), Tomlin recorded a strikeout rate of 13.2% (4.9 K/9). This season, in 86.2 innings, he’s improved upon that mark by about 50%, striking out 21.3% of all batters faced (8.0 K/9) — this, while also only walking about a batter per every nine innings. The precise reasons for his improvement are the province of another post. The effect on his projection is clear, however: the rest-of-season variety portends an improvement of 1.2 strikeouts per nine relative to the preseason one.

***

3. Tanner Roark, RHP, Washington (Profile)

Proj. IP K/9 BB/9 HR/9 FIP WAR
PRE 150 6.1 2.8 1.0 4.16 -0.1
ROS 150 6.5 2.5 0.9 4.02 1.1
Diff 0.4 -0.3 0.0 -0.14 1.2

Notes
Between 2011 and -13, Roark recorded a strikeout-walk differential slightly greater than 13 percentage points. As an actual major-league starter this season, the 27-year-old has produced a strikeout-walk differential of slightly greater than 13 percentage points. If one takes for granted that most of the hitters in the major leagues are better than most of the hitters in Double- and Triple-A — a reasonable assumption, that — than one is led to believe that a pitcher, upon graduating to the majors, will produce less impressive strikeout-walk differentials (those differentials serving as a good proxy for overall success). That hasn’t been the case for Roark, however. His rest-of-season projections reflect that.

***

2. Roenis Elias, LHP, Seattle (Profile)

Proj. IP K/9 BB/9 HR/9 FIP WAR
PRE 150 6.0 3.8 1.2 4.82 0.2
ROS 150 7.7 3.5 1.0 4.15 1.5
Diff 1.7 -0.3 -0.2 -0.67 1.3

Notes
There has been considerable and justified enthusiasm over the past year-plus regarding Seattle’s young pitchers; however, almost all of it (i.e. all that enthusiasm) has been directed in the direction of Danny Hultzen, James Paxton, and Taijuan Walker. That triumvirate have accounted for only 31.1 major-league innings this season and slightly less than zero wins relative to replacement level. The left-handed Elias, meanwhile — who received so little consideration as to be omitted from a Mariners’ consensus top-26 prospect list — has produced a league-average xFIP and 1.4 WAR in 129.0 innings. At such a time as he records two more outs, he’ll pass the injured Masahiro Tanaka for first place among rookie pitchers by innings pitched.

***

1. Dallas Keuchel, LHP, Houston (Profile)

Proj. IP K/9 BB/9 HR/9 FIP WAR
PRE 150 6.0 3.0 1.0 4.23 0.5
ROS 150 6.4 2.7 0.8 3.87 1.8
Diff 0.4 -0.2 -0.2 -0.36 1.3

Notes
Keuchel produced above-average fielding independent numbers last season (18.0% K, 7.6% BB, 55.8% GB), but those figures were such a departure from his recent minor- and major-league history that he entered 2014 with relatively modest projections. What he’s done this season, however, is to produce better fielding-independent numbers (18.7% K, 6.0% BB, 61.8% GB) and to do so exclusively as a starter (after working some of 2013 in a relief capacity). The projections are still conservative regarding his true talent, forecasting him to produce league-average numbers roughly on a rate basis; however, that’s likely product of those modest pre-2013 figures that are still extent in the data which informs his projections.

***

Two additional notes:

  • Had he not been demoted in mid-July, Arizona right-hander (and Chase Anderson’s teammate) Mike Bolsinger would actually appear atop this list. His preseason projections called for him to produce a -1.4 WAR; his current rest-of-season ones, a 0.8 mark (i.e. 2.2 wins better).
  • The least improved pitcher by this methodology appears to be Jake Peavy. Per 150 innings, his combined WAR projection has declined from 2.7 to 1.5.



Print This Post



Carson Cistulli occasionally publishes spirited ejaculations at The New Enthusiast.


25 Responses to “The Most Improved Pitchers This Season by Projected WAR”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. _David_ says:

    One tiny quibble: Hultzen wasn’t in the preseason picture, as his shoulder was already destroyed.

    Vote -1 Vote +1

  2. Rawson says:

    An interesting exercise, thanks. So, check my understanding of your findings: even though they are the most improved they still don’t project to be very good, right?

    Vote -1 Vote +1

  3. Hurtlocker says:

    So if a pitcher was projected to -3.0 WAR and actually pitched at -.5 WAR he would be the most improved?? From sucking bad to just sucking??

    Vote -1 Vote +1

  4. david says:

    I was excited when I saw this title as I am dieing for someone to explain to me how Phil Hughes has 4.0 WAR so far, more than names such as David Price, Max Scherzer & Adam Wainwright, in significantly less innings pitched. If someone could go some of the way to explaining that I would appreciate, ta

    Vote -1 Vote +1

    • Hurtlocker says:

      WAR is a mystery, surrounded by an enigma. You must have faith.

      Vote -1 Vote +1

    • Matt says:

      Besides Phil Hughes having a higher FIP than all of those guys?

      Vote -1 Vote +1

      • david says:

        Thanks Matt for the answer, it seems simple and decidedly succinct. But why is FIP actually lower? I guess he is just getting a ton of credit for barely walking anyone. But again I’d love if someone could expand on that. To me he looks like a guy who lives high in the zone with an average fastball and doesn’t have a particular out pitch leading to an earnt high BABIP & line drive rate. Ergo the low walk rate is absolutely necessary to even be starting in the majors. But more valuable to a team than Price or Wainwright?

        Vote -1 Vote +1

        • Steven says:

          FIP = ((13*HR)+(3*(BB+HBP))-(2*K))/IP + constant

          Vote -1 Vote +1

        • david says:

          Thanks Steven, guess I could of looked that up somewhere. The formula explains a lot about how guys like Hughes and Edwin Jackson show up so well with FIP.

          Vote -1 Vote +1

    • Kim says:

      That’s why you don’t use FG fir WAR. Baseball Ref has him at a much more reasonable 1.9 WAR.

      Vote -1 Vote +1

  5. mch38 says:

    I clicked on this hoping Vance Worley would fall somewhere in the top 5. Unfortunately Steamer has Worley as a .6 WAR projection prior to the season. So at .7 WAR in only 8 starts he doesn’t make the list =[ #vanimalprobs

    Vote -1 Vote +1

  6. Babyspittle says:

    Looks like Collin McHugh is right there too, according to the war value (0.4) listed at this link.
    http://www.fangraphs.com/blogs/2014-zips-projections-colorado-rockies/

    The zips update projects him at 2.1 war, which would be an increase of 1.7 WAR.

    Vote -1 Vote +1

    • Steven says:

      I think Carson is actually using the ZIPs(R) and Steamer(R) numbers rather that the updated version. The reasoning, I guess, is that the RoS projections will demonstrate how much a pitcher has improved, whereas the updated projections would demonstrate which pitcher has over performed. He was choosing to write a story on the former.

      Using this line of thinking, McHugh was projected at .5 WAR/150 innings before the season. Now he is projected at .8 zWAR/150 and 1.3 sWAR/150, or the average of ZIPs and Steamer, .9 WAR/150. Thus, he has improved by .4 WAR and did not appear on the list.

      Vote -1 Vote +1

      • Babyspittle says:

        Where are you getting the 0.8 warZ value and 1.3 warS value?
        Nevermind, I see what you did (multiplying the (R) war by a value to get to 150 innings).

        I’m sure fangraphs has already done a list of actual war values compared to pre-season projections.

        Vote -1 Vote +1

  7. joe says:

    Am I missing something? What about Garrett Richards?

    Vote -1 Vote +1

    • Babyspittle says:

      Same mistake I was making – its a different comparison.

      Measuring actual WAR to pre-season WAR, he probably comes out on top.

      Vote -1 Vote +1

      • Erik says:

        Is this essentially saying that zips/steamer don’t think his improvement is as significant or real?

        In other words his improvement in WAR over last year has more to do with under performing expectations last season and out performing expectations this season than with actual improvement?

        Reading the RoS projection on him is a little surprising. Having watched him pitch I’d say the results are real, night and day. I’m more concerned with his potential for injury than I am with him giving up runs.

        Vote -1 Vote +1

  8. GreenMountainBoy says:

    What about Garret Richards and Tyson Ross? Don’t have the stats in front of me (and don’t care frankly), but both have to be 3.0 WAR guys already this year. What were their projections? Were they THAT high that they don’t appear on this list? Something’s amiss from what my eyes tell me.

    Vote -1 Vote +1

    • Steven says:

      Again, this article is looking at the difference between a pitcher’s future projection and beginning of season projection, NOT a pitcher’s performance vs beginning projection. Richards, for example, is currently projected as a 1.8 WAR/150 inning pitcher (1.5 ZIPs and 2.0 Steamer). He began the year as a .9 WAR/150 pitcher (.4 ZIPS, 1.3 Steamer). Thus, he has improved his projection by .9 WAR, and thus did not make the list.

      Vote -1 Vote +1

  9. Flim Flam says:

    Surprised Pat Neshek is not on this list. He is at 1.4 WAR right now and hasn’t posted a positive WAR since 2008.

    Vote -1 Vote +1

  10. Eric F says:

    Carson how can you keep posting on fangraphs like nothing is happening?

    #KeepNotGraphs

    Vote -1 Vote +1

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>