2013 Pitching Rankings: Fluctuation | Disciples of Uecker

Disciples of Uecker

We'd like to go to the Playoffs, that would be cool.

There’s a line that fans can find in baseball articles, here and there, when runs scored decreases in a given season. That line goes something like, “in a year where pitching dominated,” or, a “pitching-heavy league;” the line suggests that where runs scored decreases, pitchers must have adroitly prevented those runs. The 2013 National League was one such season, as the general league runs average ((9*R)/(IP)) dropped from approximately 4.31 in 2012 to 4.04 in 2013. One might suggest, given such data, and such assumptions about pitching-heavy seasons in the past (1968, the end of the modern deadball era, comes to mind), that 2013 was a year of elite pitching.

2013 NL 100+ IP Pitching Rankings
2011-2013 NL Rotations with Runs Prevented

On the other hand, while ranking starting pitching performances by runs prevented, it becomes clear that the 2013 pitchers were not necessarily great. Certainly, Clayton Kershaw boasted one of the best ace seasons in a few years, and Jose Fernandez and Matt Harvey broke onto the scene with exceptional performances during their first full season. Yet, as sparkling at the top pitchers were, the worst pitchers were even worse in 2013 than in 2012. Ryan Vogelsong was as bad — or worse — this year than Tim Lincecum was last year, and Edinson Volquez and Barry Zito were both much worse.

Citing these three pitchers is important to note because the 2013 NL also featured five fewer starters that worked 100+ IP. In terms of basic league average, that means that the league basically lost five pitchers that could work approximately 20 starts (or more) in a season. The shortage of pitchers becomes apparent when the 2012 and 2013 pitching rankings are placed side-by-side. Here are some basic markers of runs prevented quality from both leagues:

RANK 2013 (Runs prevented) 2012 (Runs Prevented)
FIRST C. Kershaw (46) J. Cueto (38)
#5 A. Wainwright (22) J. Zimmermann (25)
#10 J. Teheran (17) M. Latos (20)
#15 M. Latos (14) S. Strasburg (14)
#20 S. Strasburg (12) M. Buehrle (9)
#25 S. Miller (10) M. Estrada (6)
#30 H. Alvarez (5) L. Harrell (4)
#35 R. Nolasco (2) M. Bumgarner (1)
#40 J. Turner (0) J. Kelly (0)
#45 T. Koehler (-6) R. Halladay (-2)
#50 M. Cain (-9) J. Westbrook (-4)
#55 J. Marquis (-13) D. Gee (-5)
#60 J. Hefner (-19) R. Nolasco (-9)
#65 E. Jackson (-27) K. Correia (-12)
LAST #68 E. Volquez (-42) #73 T. Lincecum (-33)

Having fewer pitchers work full seasons in the 2013 NL hurt clubs in at least two specific ways:

(1) The quality of pitching performances at the top of rotations dropped off faster. For example, in 2012, seven pitchers prevented between 20 and 26 runs, while five pitchers did so in 2013. The gap between the #3 and #4 rankings was five runs prevented in both leagues, but the gap between #10 and #4 was a couple runs larger in 2013 than in 2012. Oddly enough, in distributive terms, this resulted in slightly more top ranked pitchers spread around different teams, rather than concentrated on the same team. (Surely, since the Nationals and Reds rotations declined in 2013, this factored in; in 2012, those two clubs owned seven of the top 20 spots, but they earned only three of those spots in 2013).

(2) The number of pitchers that were moderately below average plummeted in 2013. Basically put, this means that the standard #3 and #4 starters that work lots of innings — albeit at a slightly-to-moderately below average rate — were almost non-existent in 2013. In last year’s National League, 24 pitchers were between 0 and 10 runs below their league and park; 18 of these pitchers were between 0 and 5 runs below average. This year, 12 starters were between 0 and 10 runs below average, with five between 0 and 5 runs below average. In 2013, seven of those 0-10 run below average starters worked 140 IP, while thirteen such starters worked in last year’s NL. One can argue the merits of using the term “innings-eater” in a positive way, but one way or the other, the innings eater really did not exist in 2013.

STARTER 2012 IP 2012 Runs Average 2013 IP 2013 Runs Average
#3 154 4.27 144.3 4.16
#4 144.3 5.05 155.3 4.92
LEAGUE 4.32 4.04

If one normalizes these pitching rankings outside of their park factors, and simply scales them to the league, the shift in the middle of the rotation is even more apparent. The biggest shift for team rotations is that #3 starters were largely below average in 2013; while #4 starters were below average in both seasons, #4 starters were worse in 2013 than in 2012 (by about 3 runs per 150 IP, or the difference between 12 and 15 runs below average).

Perhaps the most straightforward way to state it is, while five starters were at least 15 runs below average (or worse) in 2012, twelve were at least 15 runs below average (or worse) this year. In any case, it would be short-sighted to simply proclaim that the league’s runs scored declined because of elite pitching. In fact, the 2013 NL is arguably more of a bottom-heavy pitching league than a top-heavy one.

FIP Fluctuations
In 2011, I began keeping a list of pitchers with rankings and runs prevented performances that had Fielding Independent Pitching ratings that deviated from their ranking. The theory goes, if a pitcher can control things like K, BB, and HR, to a certain extent we can expect them to give up a specific number of runs (based on expected or average defensive efficiency; we could even dig deeper, use line drive, groundball, and flyball rates to form FIP expectations, or even place FIP on a park-adjusted scale for K, BB, and HR). By keeping track of FIP over a period of several years (2009-2011, 2010-2012, and 2011-2013), we can see how a pitcher’s runs allowed develop against their FIP.

2011 FIP Deviations 3YFIP 2011 IP 2011 Runs Prevented 2012 IP 2012 Runs Prevented
I. Kennedy 0.56 222 33 208.3 4
J. Cueto 0.79 156 23 217 38
T. Lincecum -0.27 217 20 186 -33
J. Chacin 0.81 194 19 69 5
J. Karsten 1.50 162.3 8 90.7
J. Saunders 1.69 212 7 130 -3
C. Carpenter 0.02 237.3 6 17 1
M. Bumgarner -0.04 204.7 6 208.3 1
M. Leake 1.27 167.7 5 179 -5
B. Beachy 0.00 141.7 3 81 15
M. Latos 0.20 194.3 2 209.3 20
Z. Greinke -0.28 171.7 -1 123 12
U. Jimenez 0.16 123 -1 AL AL
D. Moseley 1.63 120 7 Emergency Emergency
D. Gee 1.84 160.7 -13 109.7 -5
C. Billingsley 0.38 188 -14 149.7 3
J. Garcia 0.22 194.7 -14 121.7 -1
C. Volstad 1.50 165.7 -18 111.3 -28
R. Nolasco 0.46 206 -20 191 -9
E. Volquez 1.64 108.7 -21 182.7 -7

From 2011-2013, there are 212 regular starting pitching spots with 100+ IP in the National League. Among those spots, 63 have deviated from their FIP performance. This means, good or bad, that pitcher is ranked according to their runs allowed despite having a FIP that deviates from his rotation spot. For example, consider Adam Wainwright; in 2012, he was four runs below average, but he boasted a 3-year FIP of -0.19, which deviated significantly from his rotation spot for that season. One might expect Wainwright to improve, given his ability to limit walks and home runs, and sure enough, he improved notably in 2013. This year, he was 22 runs above average, and his 3-year FIP improved to -0.35. Ironically, even though Wainwright improved, his FIP is simply so good that, once again, he deviated from his rotation spot. One might say, again, that they expect Wainwright to be better, but at a certain point it becomes impractical to say that a pitcher should improve because of his FIP; some pitchers are simply so good at striking batters out, or limiting walks, or limiting HR, that their actual performance cannot mimic the strength of their FIP (Zack Greinke suffers this fate, at times).

2012 FIP Deviations 3YFIP 2012 IP 2012 Runs Prevented 2013 IP 2013 Runs Prevented
C. Kershaw -0.32 227.7 35 236 46
C. Lee -0.33 211 23 222.7 25
B. Arroyo 1.64 202 17 202 4
S. Strasburg -0.66 159.3 14 183 12
Z. Greinke 0.02 123 12 177.7 22
J. Johnson -0.27 191.3 8 AL AL
M. Fiers 0.00 127.7 8 22.3 -10
L. Lynn 0.15 176 6 201.7 -4
A.J. Burnett 1.10 202.3 5 191 2
R. Detwiler 1.09 164.3 4 71.3 -5
M. Bumgarner 0.05 208.3 1 201.3 13
K. Kendrick 1.42 159.3 1 182 -21
J. Garcia 0.13 121.7 -1 55.3 -2
R. Halladay -0.25 156.3 -2 52 -20
J. Saunders 1.38 130 -3 AL AL
A. Wainwright -0.19 198.7 -4 241.7 22
V. Worley 0.41 133 -5 AL AL
K. Correia 1.53 171 -12 AL AL
B. Zito 1.35 184.3 -13 133.3 -41
R. Wolf 1.42 142.3 -23 DNP DNP
T. Lincecum 0.33 186 -33 197.7 -23

One might simply expect fluctuation to be a part of building a rotation. Just as nearly 30% of regular starters from 2011-2013 featured FIP that deviated from their rotation spot, 40% of National League clubs have had one of their rotations rank in the Top Third and Bottom Third between 2011 and 2013. For a lot of different reasons, a significant percentage of pitching performances change from one year to the next; this could be due to injuries, bad luck, mechanical inefficiencies, changing teams, etc. We should not always expect pitchers to improve, or decline, based on their FIP; from the 2011 and 2012 deviations, eight of 41 pitchers who were expected to improve did improve, five were expected to improve and stayed the same or declined. Pitchers like Zito, Volquez, and Chris Volstad had FIP that were so bad, it’s difficult to say one way or the other if one expected them to get “better” or “worse.” Ironically, the most common performance shift, from one year to the next, among these 63 FIP deviating starters is injury, league change, or role demotion.

For next year, the pitchers that we can watch for FIP-improvement, or FIP-decline, include:

2013 FIP Deviations 3YFIP 2013 IP 2013 Runs Prevented
C. Lee -0.26 222.7 25
A. Wainwright -0.35 241.7 22
J. Chacin 0.94 197.3 22
T. Wood 1.07 200 21
K. Medlen -0.11 197 14
M. Bumgarner -0.07 201.3 13
S. Strasburg -0.28 183 12
M . Leake 1.07 192.3 10
G. Cole -0.21 117.3 7
H. Alvarez 1.19 102.7 5
B. Arroyo 1.61 202 4
T. Hudson 0.34 131.3 1
J. Turner 1.48 118
L. Lynn 0.11 201.7 -4
T. Koehler 1.23 143 -6
R. Delgado 1.52 116.3 -6
M. Cain 0.22 184.3 -9
B. McCarthy 0.26 135 -9
J. Marquis 1.74 117.7 -13
T. Chatwood 1.37 111.3 -15
B. Zito 1.66 133.3 -41

RESOURCES:
Baseball-Reference. Sports Reference, LLC., 2000-2013.
2013 NL 100+ IP Pitching Rankings
2011-2013 NL Rotations with Runs Prevented

Share Our Posts

Share this post through social bookmarks.

  • Delicious
  • Digg
  • Newsvine
  • RSS
  • StumbleUpon
  • Technorati

Comments

Tell us what do you think.

  1. joe says: November 14, 2013

    What is your formula for “runs prevented”. Does it take into account the defense, the ball park, the catcher, the situation of the game, etc.?

    • Nicholas Zettel says: November 15, 2013

      Good question, Joe.

      I try to keep runs prevented as basic as possible, so I calculate it this way:

      (1) First, I use league runs average ((9R)/(IP)).
      (2) I multiply league runs average by a three-year park factor for each team (“PARK”).
      (3) I use that three-year factor to determine the number of runs each pitcher was expected to allow ((Individual IP)/(9))*(PARK) = “EXPECTED RUNS”
      (4) Then, I subtract the pitcher’s RUNS from EXPECTED RUNS (EXPECTED RUNS – INDIVIDUAL RUNS)

      This will automatically account for team defense, since it uses runs allowed. I do not count defense individually, because defensive efficiency is not typically distributed evenly between pitchers. Game situations are included in the basic season performance; no special situational bonuses or demerits added. Catcher is not individually accounted for, he is a part of team defense.

      Hope that helps! At base, “runs prevented” is a basic way of judging a pitcher against his environment. Thanks for reading and commenting.

Trackbacks

Websites mentioned my entry.

  1. Resource: Fun with FIP, 2011-2013 | Disciples of Uecker

Add a Comment

Fill in the form and submit.