Which pitchers had the biggest differences last year between their actual ERA & their FIP ERA? (or Bad Luck is an Aaron Bummer)

FIP ERA means fielding independent ERA and it is an estimated ERA based on what the pitcher controls: walks, strikeouts and HRs. See Baseball By The Numbers—Earned Run Average (ERA) and Fielding Independent Pitching (FIP) by Marilyn Green at "Redbird Rants" for more information and the formula.

I used Stathead from Baseball Reference to call up all the pitchers who had 50+ IP last year. Then I found the difference between their actual ERA and their FIP ERA.

Table 1 shows the guys who had the worst luck, that is they had the biggest negative differentials for FIP ERA - ERA.

Table 1

Player	IP	ERA	FIP	Diff
Aaron Bummer	58.1	6.79	3.58	-3.21
Shintaro Fujinami	79	7.18	4.61	-2.57
Zach Davies	82.1	7.00	4.58	-2.42
Hogan Harris	63	7.14	5.02	-2.12
Fernando Cruz	66	4.91	2.83	-2.08
Dylan Floro	56.2	4.76	2.96	-1.80
Michael Grove	69	6.13	4.36	-1.77
Connor Seabold	87.1	7.52	5.75	-1.77
Osvaldo Bido	50.2	5.86	4.10	-1.76
Josh Sborz	52.1	5.50	3.75	-1.75

Table 2 shows the guys who had the best luck, that is they had the biggest positive differentials for FIP ERA - ERA.

Table 2

Player	IP	ERA	FIP	Diff
Wandy Peralta	54	2.83	5.05	2.22
Héctor Neris	68.1	1.71	3.83	2.12
Tom Cosgrove	51.1	1.75	3.70	1.95
Brusdar Graterol	67.1	1.20	3.03	1.83
Kendall Graveman	66.1	3.12	4.88	1.76
Dominic Leone	54	4.67	6.29	1.62
Clayton Kershaw	131.2	2.46	4.03	1.57
Wade Miley	120.1	3.14	4.69	1.55
Bryse Wilson	76.2	2.58	4.13	1.55
Ronel Blanco	52	4.50	5.99	1.49

Bummer's differential is much greater than anyone else's. I thought that maybe he got hit hard with runners on base, but his splits don't reflect that. Did the White Sox have bad fielding? Maybe, their team FIP was 4.71 while the team ERA was 4.87. Bad, but not too bad. Certainly not close to what happened to Bummer. And if it was the fielders, we would see high numbers for his AVG & SLG allowed with runners on base, but again, that was not the case (that would not be the whole story but at least part of it). His splits can be found easily at BR.

Maybe the guys who came in after he had put some runners on just did really bad so that he got charged for the runs. I am not sure if there is an easy way to tell that.

Others have looked at the issue of what explains the difference between FIP and ERA before. I might post some of those links when I get a chance.

Update Feb. 22: Here are Bummer's AVG-OBP-SLG for his splits last year:

None on)  .240-.365-.365

Runners on) .233-.342-.333

 

Looks like he had good numbers with runners on. He did allow .340 BAbip while his AVG allowed in all situations was .236, a .104 difference. For all of the AL last year those numbers were .295/.245. For the NL they were .299/.251. So Bummer's differential here was twice the MLB average. That might partly explain why his FIP/ERA differential is so large.

 

Here are some studies on this issue:

 

The FIP/ERA Gap: Historically by Glenn DuPaul of Beyond the Box Score.

 

Evaluating the Gap Between ERA and FIP by Christopher Rinaldi of FanGraphs.