Dominican League Performance as Indicator of Future Value

At the end of October, the present author published a not entirely exhaustive study which considered to what degree success in the Arizona Fall League (AFL) either did or did not portend future major-league success. Specifically, I looked at those AFL pitchers who’d finished either among the top or bottom third of that league’s starters by strikeout rate — the justification being that strikeout rate (a) is not only a fairly reliable indicator of success, but also (b) becomes stable in relatively small samples.

Even more specifically, I did this:

To better understand how much pitcher strikeout rate in the AFL rate might inform future major-league success, I looked at AFL pitchers from 2005 to -09 who’d both (a) faced 70 or more batters (that is, the sample threshold at which strikeout rate becomes reliable for major-league pitchers) and also (b) made at least half their appearances as starts. Each year in the DWL there are about 25-30 pitchers who meet both those criteria. For each of the five years in question, I isolated both the top and bottom third by AFL strikeout rate, resulting in 43 “high strikeout” and 43 “low strikeout” pitchers (although it should be noted that Adam Bostick appears twice among the high-strikeout group, thus leaving the former group with just 42 different actual players).

Despite the fact that Arizona Fall League performance is largely ignored — or, when it is cited, couched in a surfeit of qualifiers — the results of the study indicate that those pitchers who perform well in the AFL are more likely to perform well in the majors. By a reasonable margin, it seems. Again, the findings hardly suggest that AFL performance ought to be utilized without any context at all, but merely that — so far as pitchers are concerned, at least — it appears to provide non-negligible information regarding future value.

Of course, the AFL isn’t the only extant fall or winter league. Indeed, a number of other players — prospects as well as veterans and journeymen — all participate in the various Caribbean Leagues, the playoffs for which are currently underway in their respective countries. “What?” I asked myself in the service of ignoring something much darker and foreboding in my life, “what might winter-league performance say or not say about a particular pitcher’s future value?”

To answer that question — or at least begin to answer it — I utilized a methodology not unlike that one noted in the AFL study mentioned above (the justification for which methodology is explained in greater depth within that original post). For the earliest five years for which Dominican Winter League (DWL) data is publicly available — in this case, from 2007 to -11 — I looked at DWL pitchers who’d both (a) faced 70 or more batters* and also (b) made at least half their appearances as starts. Each year in the AFL there are about 30 pitchers who meet both those criteria. For each of the five years in question, I isolated both the top and bottom third by DWL strikeout rate, resulting in 51 “high strikeout” and 51 “low strikeout” player seasons.

*While Russell Carleton identifies 70 batters faced as the sample at which strikeout rate stabilizes, that threshold also oughtn’t be regarded as a “magic number,” but rather (according to Carleton) “the point where a measure of reliability slowly crosses an only-somewhat-arbitrary line in the sand.”

I say “player seasons” because this method of isolating high- and low-strikeout groups doesn’t actually produce two separate samples of 51 discrete players each. Unlike with the AFL, it’s common for players (especially those native to the Dominican) to participate in the DWL over the course of multiple seasons. Right-hander Ramon Ortiz, for example, meets the criteria for inclusion in this study in each of the five years between 2007 to -11. As a result, there’s no means by which to ensure a totally even distribution of high- and low-strikeout pitchers. The inability to do so, however, oughtn’t alter the results significantly. In the end, one is left with 41 distinct “high strikeout” and 34 distinct “low strikeout” pitchers for the purposes of this study.

For reference, during their respective DWL seasons, the high-strikeout group produced a collective 22.3% strikeout rate; the low-strikeout group, a figure of just 11.1%.

Having arrived at the two distinct samples, I then utilized the leaderboard feature here at the site to find the major-league outcomes for all the relevant pitchers since 2008 — that is, the first full major-league season major-league following the 2007-08 DWL season which serves as the first of the five DWL years in this study.

Here are the 28 members of the high-strikeout group (out of 41 total) to have recorded at least one major-league appearance during that interval, sorted by RA9-WAR (that is, WAR calculated with runs allowed per inning innings, as opposed to FIP):

#	Name	Team	G	GS	IP	K%	BB%	xFIP-	FIP-	ERA-	WAR	RA9-WAR
1	Johnny Cueto	Reds	194	194	1208.1	19.9%	7.3%	97	95	80	17.0	26.1
2	Ivan Nova	Yankees	93	86	537.2	17.5%	7.8%	98	101	101	6.3	7.4
3	Alfredo Simon	– – –	209	51	529.1	16.6%	7.5%	107	112	97	1.2	4.3
4	Wily Peralta	Brewers	70	69	411.0	17.5%	8.3%	103	107	101	3.4	3.4
5	James McDonald	– – –	131	82	512.1	20.0%	10.4%	112	111	110	3.2	3.1
6	Robinson Tejeda	– – –	127	7	187.1	24.0%	13.0%	103	88	86	2.3	2.1
7	Fernando Abad	– – –	196	6	179.2	19.4%	7.6%	110	107	94	0.5	1.4
8	Dustin Nippert	Rangers	78	18	198.0	17.3%	11.1%	112	105	108	1.2	1.4
9	Rafael Perez	Indians	276	0	256.0	17.3%	8.7%	90	90	96	2.0	1.3
10	Kyle Davies	Royals	88	88	481.0	15.6%	10.1%	113	107	121	4.8	1.3
11	Claudio Vargas	– – –	64	4	98.0	16.9%	8.9%	106	103	96	0.3	1.1
12	Aquilino Lopez	Tigers	48	0	78.2	17.7%	6.4%	107	91	81	0.7	1.0
13	Everett Teaford	Royals	45	8	106.0	14.3%	7.9%	109	127	103	-0.5	0.8
14	Scott Lewis	Indians	5	5	28.1	15.4%	6.0%	110	125	81	0.1	0.6
15	Roberto Hernandez	– – –	178	167	975.0	13.8%	9.0%	108	119	122	3.8	0.3
16	Wesley Wright	– – –	361	4	299.2	22.1%	10.1%	95	107	104	-0.4	0.2
17	Joselo Diaz	Rangers	1	0	1.0	33.3%	16.7%	149	115	0	0.0	0.0
18	Jose Capellan	Rockies	1	0	2.0	22.2%	0.0%	41	24	97	0.0	0.0
19	Doug Mathis	Rangers	45	6	87.1	11.5%	9.1%	110	120	107	-0.2	0.0
20	Evan MacLane	Cardinals	2	0	1.0	0.0%	25.0%	182	488	230	-0.2	-0.1
21	Yunesky Maya	Nationals	16	10	59.0	10.3%	8.0%	126	127	149	0.0	-0.3
22	Rafael Rodriguez	– – –	21	0	35.1	7.7%	7.1%	111	127	128	-0.3	-0.4
23	Raul Valdes	– – –	103	3	140.1	24.5%	8.2%	102	105	134	0.5	-0.8
24	Aneury Rodriguez	Astros	44	9	91.1	17.9%	8.7%	122	126	135	-0.1	-0.9
25	Chris Leroux	– – –	65	0	71.2	19.8%	10.5%	99	85	153	0.7	-1.3
26	Radhames Liz	Orioles	19	17	85.2	14.2%	13.0%	135	144	171	-0.3	-1.4
27	Esmil Rogers	– – –	192	43	421.0	18.7%	8.7%	103	105	129	2.5	-1.6
28	Jorge Sosa	– – –	60	2	80.2	12.8%	10.9%	129	133	140	-0.9	-1.9

And here are the 15 members of the low-strikeout group (out of 34 total) to have recorded at least one major-league appearance over that time period:

#	Name	Team	G	GS	IP	K%	BB%	xFIP-	FIP-	ERA-	WAR	RA9-WAR
1	Brandon McCarthy	– – –	119	119	736.0	16.9%	4.8%	92	90	97	12.5	9.5
2	Greg Smith	– – –	40	40	229.1	14.5%	11.3%	121	118	105	1.6	2.8
3	Francisley Bueno	– – –	56	0	60.1	12.9%	4.7%	109	90	76	0.5	1.0
4	Luis Perez	Blue Jays	78	4	112.0	20.2%	9.2%	93	98	108	0.6	0.2
5	Adalberto Mendez	Marlins	5	5	24.2	10.1%	11.0%	154	186	125	-0.5	0.1
6	Jose Capellan	Rockies	1	0	2.0	22.2%	0.0%	41	24	97	0.0	0.0
7	Henry Sosa	Astros	10	10	53.1	16.5%	10.0%	117	125	138	0.1	-0.1
8	Josh Banks	– – –	24	18	112.0	10.7%	8.1%	122	141	142	-1.0	-0.5
9	Cesar Valdez	D-backs	9	2	20.0	13.4%	10.3%	113	112	180	0.1	-0.6
10	Frankie de la Cruz	– – –	20	1	25.1	12.0%	17.6%	151	140	203	-0.3	-0.6
11	Hector Noesi	– – –	97	48	362.2	16.4%	8.2%	114	124	131	0.1	-0.8
12	Ramon Ortiz	– – –	45	8	88.2	13.7%	9.7%	123	146	144	-0.9	-1.1
13	Lenny DiNardo	– – –	16	7	44.1	8.7%	12.1%	126	134	200	-0.2	-1.5
14	Jorge Sosa	– – –	60	2	80.2	12.8%	10.9%	129	133	140	-0.9	-1.9
15	Julian Tavarez	– – –	94	0	89.2	19.0%	12.6%	101	97	118	0.1	-2.0

Some observations from the data:

• Among the Dominican League’s high-strikeout group, 28 of 41 pitchers (68%) have recorded at least one major-league appearance since 2008. Among the low-strikeout group, only 14 of 34 pitchers (41%).

• Among the high-strikeout group, 16 of 41 pitchers (39%) have produced a positive WAR figure in the majors since 2008. Among the low-strikeout group, only 5 of 34 pitchers (15%).

• Among the high-strikeout group, 5 of 41 pitchers (12%) have produced a strikeout rate above 18% (i.e. the league-average rate since 2008) while working as a starter in the majors (minimum five starts). Meanwhile, 0 of 34 pitchers (0%) from the low-strikeout group have done the same.

• The median pitcher from the high-strikeout group has produced -0.3 RA9-WAR since 2008. The median pitcher from the low-strikeout group hasn’t appeared in the majors, but would (theoretically) have produced something worse than Julian Tavarez’s -2.0 RA9-WAR.

• Here are the top-10 DWL seasons by strikeout rate from 2007 to -11 among pitchers who faced at least 70 batters and registered at least half their appearances in a starting capacity:

#	Player	Year	IP	TBF	K%
1	Everett Teaford	2012	23.2	91	32.0%
2	Fabio Castro	2010	51.0	212	28.8%
3	Joselo Diaz	2008	22.1	99	28.2%
4	Johnny Cueto	2007	31.2	132	28.1%
5	Hiroshi Kisanuki	2008	19.0	80	27.5%
6	Scott Lewis	2008	22.0	97	26.8%
7	Nerio Rodriguez	2010	43.1	176	26.7%
8	Yunesky Maya	2010	41.0	159	26.4%
9	Kenji Otonari	2012	23.0	103	26.2%
10	Aneury Rodriguez	2012	32.2	125	25.7%

And here, for some reason, are the 10 worst such seasons by that same criteria:

#	Player	Year	IP	TBF	K%
1	Jose Mercedes	2010	17.0	73	4.1%
2	Lindsay Gulin	2009	17.2	80	5.0%
3	Hector Carrasco	2009	20.2	101	7.0%
4	Josh Banks	2010	37.0	157	7.0%
5	Sam Narron	2008	14.1	70	7.1%
6	Sam Narron	2009	41.2	182	7.7%
7	Jose Lima	2009	31.0	141	7.8%
8	Andy Van Hekken	2010	19.2	86	8.2%
9	Ramon Ortiz	2007	19.1	79	8.8%
10	Jose Capellan	2009	25.1	124	8.8%

Notably, two of the best 10 DWL seasons between 2007 and -11 were produced by Japanese pitchers. Both have remained active in the NPB through 2014.

*****

Conclusions
As was the case with the Arizona Fall League, it appears as though pitchers who produce high-ish strikeout rates in the Dominican Winter League are more likely than those who record low-ish ones both to (a) make a major-league appearance of some sort in subsequent years and (b) record a positive WAR figure in the majors in subsequent years. Success of this sort in the DWL does not appear to guarantee success in the majors. (The median sort of pitcher from the high-strikeout group has basically been replacement level in subsequent major-league seasons.) That said, it’s a more positive indicator of future value than finishing among the DWL’s lowest-strikeout group.

Finally, below are the 12 pitchers who comprise the high-strikeout group from the 2014-15 Dominican Winter League season. Pitchers are listed with 2015 age and club. Again, this is the sort of pitcher who went on to make a major-league appearance in 68% of cases and to produce a positive future WAR in 39% of cases.

#	Player	Org	Age	G	GS	GS%	IP	TBF	K%
1	Radhames Liz	Pirates	31	5	5	100.0%	23.2	92	31.7%
2	Randall Delgado	D-backs	25	7	7	100.0%	33.0	138	29.7%
3	Miguel Almonte	Royals	22	4	3	75.0%	19.1	73	25.9%
4	Chris Leroux	Yankees	31	5	4	80.0%	16.1	70	25.6%
5	Asher Wojciechowski	Astros	26	5	5	100.0%	20.1	87	25.2%
6	Liam Hendriks	Blue Jays	26	6	6	100.0%	33.1	133	24.0%
7	Andre Rienzo	Marlins	27	6	5	83.3%	18.0	84	23.8%
8	Nick Additon	Brewers	27	6	4	66.7%	20.0	82	23.2%
9	Merrill Kelly	Korea	26	6	6	100.0%	22.0	103	22.3%
10	Mitch Atkins	Rangers	29	11	11	100.0%	46.0	193	21.8%
11	Amaury Rivas	N/A	29	7	4	57.1%	28.1	111	21.6%
12	Sugar Ray Marimon	Braves	26	5	5	100.0%	23.0	98	21.4%