Fast Starting Pitchers: ERA vs xFIP and Sustainability

When someone like Josh Collmenter blows through the league on his first go 'round, everyone asks the same series of questions.

What's his secret? Can he keep it up? OK, I know he can't keep it up, but what's his true level of performance?

We always talk about fast starters and how unlikely it is that they'll "keep it up." But how true is this premise?

Highest FIP/xFIP differentials, April 2011

#	Player	'11 April ERA	'11 April xFIP	Diff	'11 May ERA	'11 May xFIP	Dif
1	Alexi Ogando	1.81	3.75	1.94	1.27	3.37	2.10
2	Wade Davis	3.47	5.29	1.82	4.25	5.94	1.44
3	Jair Jurrjens	1.80	3.43	1.63	2.25	3.41	1.16
4	Zach Britton	2.14	3.74	1.60	1.32	3.74	2.42
5	Kyle Lohse	2.06	3.61	1.55	2.50	4.17	1.67
6	Josh Tomlin	2.41	3.96	1.55	2.36	4.02	1.66
7	Josh Beckett	1.73	3.25	1.52	0.39	3.69	3.30
8	Josh Johnson	1.64	3.10	1.46	3.26	3.82	0.56
9	Trevor Cahill	1.79	3.06	1.27	1.67	3.13	1.46
10	Charlie Morton	2.62	3.89	1.27	2.05	3.62	1.57

The process of putting this chart absolutely blew me away. Only Johnson came close to being luck neutral in May. The other nine pitchers in this chart are in the top 24 in ERA/xFIP differential (out of 122 qualifying pitchers in May).

I figured that some of April's lucky duckies would have turned into pumpkins in May. But it didn't happen. Not only did it not happen, but everyone save for Johnson was an entire earned run or more better than his xFIP would have suggested.

Is there a common thread with these pitchers?

Highest ERA/xFIP April differentials w/YTD rate stats

#	Player	K/9	BB/9	HR/9	BABIP	LOB%	GB%
1	Alexi Ogando	6.49	2.11	0.91	.199	92.4%	35.5%
2	Wade Davis	4.11	4.11	0.95	.257	81.5%	33.3%
3	Jair Jurrjens	5.22	1.44	0.36	.268	83.7%	50.6%
4	Zach Britton	5.03	2.90	0.61	.227	83.3%	55.6%
5	Kyle Lohse	4.96	1.70	0.36	.222	79.9%	45.7%
6	Josh Tomlin	4.53	1.51	1.21	.175	85.4%	37.0%
7	Josh Beckett	8.16	2.67	0.47	.224	86.4%	41.2%
8	Josh Johnson	8.35	2.98	0.30	.239	82.2%	51.0%
9	Trevor Cahill	7.16	2.76	0.55	.257	86.1%	56.4%
10	Charlie Morton	4.75	4.25	0.33	.257	79.5%	61.9%

Top 10 in category in BOLD

In terms of positive benchmarks, the largest commonality is luck. The median BABIP thus far this year among qualifiers is .288; Jurrjens is the only pitcher who even comes close. Likewise, the median LOB% among qualifiers is 73.5%. Lohse and Morton are the only two pitchers who even clock in below 80%.

But none of this answers the important question for Rotisserie/fantasy players. Intuitively, we feel like these rates aren't sustainable and that these players are going to regress. However, if you "sold low" on Ogando in late April you're probably not happy about it at the moment. Is there a magic moment at some point in the season where you should truly sell on these types of guys?

2010 ERA/xFIP: Fast Starters and ERA/xFIP Differentials by Month

#	Player	Apr	May	June	July	Aug	Sept	Total
1	Livan Hernandez	3.95	1.33	-0.50	0.23	0.76	0.01	0.91
2	Mike Pelfrey	3.46	-0.09	0.44	-4.40	2.67	0.59	0.65
3	Ubaldo Jimenez	2.67	2.51	-0.57	-1.67	0.59	-0.06	0.72
4	Mitch Talbot	2.63	-0.07	0.50	-0.51	-1.24	1.34	0.44
5	Jaime Garcia	2.54	2.09	-0.63	0.90	1.25	-2.26	0.92
6	Barry Zito	2.40	0.76	-0.11	1.33	-2.49	-0.20	0.43
7	Brad Penny	2.33	-2.81	N/A	N/A	N/A	N/A	0.40
8	Francisco Liriano	2.33	-2.15	-1.43	-0.38	-0.34	-1.33	-0.67
9	Doug Fister	2.21	1.09	-4.45	-1.18	-0.91	-0.78	-0.01
10	Carlos Silva	2.18	-0.74	1.11	-2.98	-35.6	-7.50	0.47

Not only isn't there a magical "sell high" point, but last year's fast starts show that this year's sustainability among the ERA/xFIP outliers is a mirage. U-Ball and Garcia managed to keep the great times rolling into May 2010, while Livan and Fister managed to do the same to some degree. But there are only three months from May-September where one of these pitchers posted a 2+ ERA/xFIP differential to the good: Jimenez May 2010, Garcia May 2010, and Pelfrey August 2010.

There is definitely room for discussion around the idea that some pitchers will defy their xFIP and pitch to a better ERA based on their skill sets. However, since 2002, no pitcher with greater than 750 IP from 2002-2010 has put up a differential over 1.0, and only six pitchers have put up a differential greater than 0.75. If you're currently riding Alexi Ogando or Zach Britton and think that this is the year and my pitcher is the one who is going to finally defy the odds, well, I've got some bad news for you...