Saturday, February 08, 2014

Tigers Pitchers Got Lots of Easy Outs in 2013

 More than a third of the batters facing Max Scherzer in 2013 were "easy outs". 
 (Photo credit: Star Tribune)

Much has been written about fielding independent events - strikeouts, walks, hit batsmen and home runs - over the last several years.  These are things which a pitcher controls with minimal contribution from fielders.  No event is totally independent of teammates of course.  For example, an outfielder can reach over the fence to prevent a home run.  Additionally, recent research on framing pitches suggest that catchers can have some influence over strikeouts and walks.  Still, pitchers do control these events for the most part.

Another event which is mostly the responsibility of the pitcher, and could be considered fielding independent, is the infield fly.  While there are certainly fielders involved in getting outs on infield flies, there is not a lot of difference among major league infielders in their ability to catch infield flies.  Regardless of the pitcher or the team, when you see a pop up in the infield it almost always leads to an out.  Table 1 below shows that batters hit .022 and slugged .028 on pop ups in 2013.   These rates are much lower than those for any other batted ball type including ground balls.

Table 1: Batting Statistics by Batted Ball Type, 2013
Batted Ball Type
BA
SLG
OPS
Ground ball
.244 .262 .506
Infield fly
.022 .028 .050
Outfield fly
.231 .673 .905
Line drive
.674 .977 1.650
Data source: Retrosheet.org

The notion that infield flies are essentially fielding independent events is not not a new one.  Dave Cameron and Tom Tango have both suggested adding infield flies to the Fielding Independent Pitching (FIP) statistic similar to how strikeouts are already included.  It's true that infield fly ball rate is less predictive of future performance than strikeout rate, but predicting the future is not the only function of FIP.  It is also used to evaluate past performance.

Since infield flies lead to outs almost as frequently as strikeouts, I have combined the two into a statistic I call Easy Out Percentage - EO% (not to be confused with OE%).  It is simply strikeouts plus infield flies divided by batters faced.  As mentioned above, this could also be incorporated into FIP, but simple statistics are good too.  Table 2 below shows that Rangers ace right hander Yu Darvish led the majors with a 38.3 EO% in 2013.  The National League leader was Matt Harvey of the Mets (35.4).

Three Tigers finished in the top six - Cy Young award winner Max Scherzer (34.9), AL ERA leader Anibal Sanchez (34.7) and the "disappointing" Justin Verlander (32.8). 

Table 2: Easy Out Percentage (Strikeouts and Infield Flies), 2012

Pitcher
Team
BFP
K
IFFB
EO%
Yu Darvish
Rangers
841
277
45
38.3
Matt Harvey
Mets
690
191
53
35.4
Max Scherzer
Tigers
836
240
52
34.9
Anibal Sanchez
Tigers
746
202
57
34.7
Madison Bumgarner
Giants
803
199
65
32.9
Justin Verlander
Tigers
925
217
86
32.8
Felix Hernandez
Mariners
822
216
48
32.1
Clayton Kershaw
Dodgers
908
232
59
32.0
Jose Fernandez
Marlins
681
187
31
32.0
Chris Sale
White Sox
866
226
48
31.6
Stephen Strasburg
Nationals
731
191
39
31.5
Shelby Miller
Cardinals
722
169
57
31.3
Jose Quintana
White Sox
832
164
95
31.1
Erik Bedard
Astros
663
138
67
30.9
Ubaldo Jimenez
Indians
777
194
45
30.8
Julio Teheran
Braves
774
170
68
30.7
Cole Hamels
Phillies
905
202
75
30.6
A.J. Griffin
Athletics
823
171
80
30.5
Lance Lynn
Cardinals
856
198
63
30.5
Gio Gonzalez
Nationals
819
192
56
30.3
                         Data source: FanGraphs.com 

The majority of pitchers rank similarly on K% and EO%, but there are some exceptions.  One such outlier was Jose Quintana of the White Sox who ranked 49 out of 109 pitchers (with 600+ batters faced) in strikeout percentage, but 13th in EO%.   The reason for the disparity was that Quintana (15.6%) ranked first in infield flies per batted ball even though he whiffed relatively few batters.

Another oddball was Pirates right hander A.J Burnett - 85th in K% and 39th in EO%.  In his case, he retired a lot of batters on strikes but had the lowest pop fly rate in the majors.

It is important to keep in mind that the ability to to induce popups has not been shown to be a highly repeatable skill for most pitchers.  For example, Quintana's rate in 2013 was about twice as high as in his rookie campaign in 2012.  Regardless, his infield flies in 2013 added value and he should get credit for them.   

4 comments:

  1. Nice post. A few thoughts come to mind. First, Is the variability of IFFB rate among pitchers similar to K rate or is it a more constant figure? Second, how are foul outs handled in these statistics? Could they be a third class of Easy Out? Do they occur frequently enough to consider? Finally, knowing the league statistics by batted ball type, is there a stat that takes these league averages and applies them to the type of batted balls a pitcher has thrown? For example, if Verlander has the following (217 K/925 BFB x .000 = 0), (86 IFFB/925 BFB * .022 = .002), ( 207* GB/925 * .244 = .055), (207* OFFB/925 * .231 = .052), (208* LD/925 * .674 = .152). * Actual stat not known. 0+.002+.055+.052+.152 = .261 Would that calculated .261 BAA be a useful fielding independent (fielding actually is changed to league average)? What if you did that calculation for every BFB for a team? Would the difference between the calculated stat and the actual stat be a reasonable measure of a team's overall defensive effectiveness?

    ReplyDelete
  2. Catchy sounding stat!! :)

    Now we just need to crack the code for what is an EIEIO%?

    ReplyDelete
  3. K% has a higher standard deviation than iff% but that's because it's generally a higher number. Taking magnitude into consideration (coefficient of variation), they are similar in dispersion.

    Foul pops are included.

    If you calculate the expected batting average against as you proposed and compared it to actual batting average, it could tell you something about defense at the team level. It probably wouldn't work too well for individual pitchers due to small sample sizes. Some of the defensive stats do something like that as part of their calculation. At their core, defensive stats are all about expected outs versus actual outs.

    ReplyDelete
  4. I'll add that the Hardball Times used to have very similar stat as their +/- defensive measure. They excluded home runs from fly balls and line drives, because of those can not be handled by fielders.

    ReplyDelete

Twitter

Blog Archive

Subscribe

My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter