One inning wonders do not pitch every day.

Am I the only who thinks the current conventional wisdom of shortening games through dominant one inning relief pitchers is silly?

Are relief pitchers becoming too dominant? Mike Marshall 1974: 208 innings, 106 games, all in relief. Sunday, September 14, 2014

It seems like every pitcher who comes out of the bullpen averages more than one strike out per inning.  Actually it's only about half the relief pitchers...

Mike Marshall in 1974  won the Cy Young award: 208 innings in 106 relief appearances, SO9 6.18. ..

For some perspective: Mariano Rivera pitched 1995-2013.  Rivera threw 100 innings only once: 107 in 1996 when he was the setup man for John Wetteland.  That was also Rivera's best SO9: 10.6.  Rivera had SO9 better than 9 in only 6 of his 18 full seasons; low: 1998: SO9 5.3 in 61 innings.  Rivera's second most innings: 80 in 2001: SO9 9.3.  Rivera was under 70 innings in 8 of 18 seasons.  Rivera might be obsolete one year after retiring.
_______________________________________

In 1996 John Wetteland led the league in saves with 46 but pitched only 62.66 innings. Mariano Rivera did the heavy lifting with those 107 innings. At the time it was said that they reduced it to a seven inning game. In 2014 it's been said that some teams, most notably the Kansas City Royals, have reduced it to a six inning game.  Kansas City was 11 of 15 American Conference teams in attendance.

Rk	Player	IP	GS	Year	Age	Tm	Lg	G	CG	SHO	GF	W	L	W-L%	SV	H	R	ER	BB	SO	ERA	FIP	ERA+	HR	BF	AB	2B	3B	IBB	HBP	SH	SF	GDP	SB	CS	PO	BK	WP	BA	OBP	SLG	OPS	OPS+	Pit	Str
1	Dellin Betances	90.0	0	2014	26	NYY	AL	70	0	0	8	5	0	1.000	1	46	15	14	24	135	1.40	1.64	277	4	341	308	9	1	1	4	2	3	1	12	3	1	1	2	.149	.218	.224	.442	26	1365	891
2	Dan Otero	86.2	0	2014	29	OAK	AL	72	0	0	14	8	2	.800	1	80	24	22	15	45	2.28	3.28	163	4	348	325	12	1	7	2	4	3	10	3	3	0	0	1	.246	.281	.326	.607	72	1218	822
3	Seth Maness	80.1	0	2014	25	STL	NL	73	0	0	17	6	4	.600	3	77	29	26	11	55	2.91	3.38	126	7	317	295	12	1	3	2	5	4	12	1	3	0	1	2	.261	.288	.380	.668	90	1094	742
4	Adam Warren	78.2	0	2014	26	NYY	AL	69	0	0	11	3	6	.333	3	63	27	26	24	76	2.97	2.89	130	4	324	288	17	2	1	3	5	4	7	1	2	0	0	4	.219	.282	.333	.615	73	1333	843
5	Tony Watson	77.1	0	2014	29	PIT	NL	78	0	0	3	10	2	.833	2	64	16	14	15	81	1.63	2.69	219	5	305	276	8	2	0	6	5	3	8	4	2	0	0	0	.232	.283	.330	.613	77	1199	805
6	Jeurys Familia	77.1	0	2014	24	NYM	NL	76	0	0	16	2	5	.286	5	59	26	19	32	73	2.21	3.07	158	3	322	282	11	2	5	2	4	2	7	7	0	0	0	9	.209	.292	.294	.587	74	1221	763
7	Zach Britton	76.1	0	2014	26	BAL	AL	71	0	0	49	3	2	.600	37	46	17	14	23	62	1.65	3.13	230	4	285	258	7	0	0	1	3	0	13	2	0	0	0	0	.178	.248	.252	.500	42	1053	673
8	Bryan Shaw	76.1	0	2014	26	CLE	AL	80	0	0	16	5	5	.500	2	61	26	22	22	64	2.59	3.42	142	6	313	282	11	1	4	2	5	2	3	11	2	0	1	4	.216	.276	.326	.602	72	1247	806
9	Todd Redmond	75.0	0	2014	29	TOR	AL	42	0	0	14	1	4	.200	1	73	33	27	27	60	3.24	3.56	121	5	314	278	21	1	6	2	2	5	8	7	6	0	0	2	.263	.327	.399	.726	104	1205	773
10	Joe Smith	74.2	0	2014	30	LAA	AL	76	0	0	26	7	2	.778	15	45	16	15	15	68	1.81	2.85	202	4	285	261	6	2	3	6	3	0	8	3	0	0	0	4	.172	.234	.257	.491	42	1105	742
11	Jake Petricka	73.0	0	2014	26	CHW	AL	67	0	0	33	1	6	.143	14	67	24	24	33	55	2.96	3.60	131	3	307	265	13	0	4	2	3	4	12	2	4	1	0	2	.253	.336	.336	.671	91	1191	746
12	Craig Stammen	72.2	0	2014	30	WSN	NL	49	0	0	15	4	5	.444	0	78	34	31	14	56	3.84	3.19	99	5	304	283	14	2	2	3	3	1	8	5	1	0	1	1	.276	.316	.392	.708	99	1139	740
13	Luke Gregerson	72.1	0	2014	30	OAK	AL	72	0	0	17	5	5	.500	3	58	20	17	15	59	2.12	3.24	176	6	284	264	14	0	3	1	3	1	6	10	0	0	0	6	.220	.263	.341	.604	71	1006	665
14	Wade Davis	72.0	0	2014	28	KCR	AL	71	0	0	11	9	2	.818	3	38	8	8	23	109	1.00	1.19	403	0	279	252	3	2	0	3	0	1	3	4	1	0	0	1	.151	.229	.179	.408	16	1210	787
15	Scott Atchison	72.0	0	2014	38	CLE	AL	70	0	0	11	6	0	1.000	2	60	24	22	14	49	2.75	3.08	134	4	280	264	14	2	4	0	1	1	12	4	2	0	0	1	.227	.265	.341	.606	73	1037	667
16	Shawn Tolleson	71.2	0	2014	26	TEX	AL	64	0	0	10	3	1	.750	0	56	23	22	28	69	2.76	4.23	142	10	296	262	9	1	5	1	2	3	5	4	2	0	0	4	.214	.289	.370	.659	84	1166	708
17	Jake McGee	71.1	0	2014	27	TBR	AL	73	0	0	31	5	2	.714	19	48	15	15	16	90	1.89	1.73	197	2	274	254	8	0	1	2	1	1	1	2	1	1	0	1	.189	.242	.244	.486	40	1151	764
18	Danny Farquhar	71.0	0	2014	27	SEA	AL	66	0	0	22	3	1	.750	1	58	23	21	22	81	2.66	2.86	137	5	290	262	10	0	1	4	1	1	4	1	3	1	2	6	.221	.291	.317	.607	77	1139	735
19	Mark Melancon	71.0	0	2014	29	PIT	NL	72	0	0	48	3	5	.375	33	51	15	15	11	71	1.90	2.09	188	2	277	261	5	0	1	3	1	1	0	1	1	0	0	3	.195	.236	.238	.473	39	975	669
20	Jake Diekman	71.0	0	2014	27	PHI	NL	73	0	0	19	5	5	.500	0	66	36	30	35	100	3.80	2.65	98	4	313	266	13	2	5	3	2	7	4	5	3	1	0	7	.248	.334	.357	.692	100	1316	798
21	Burke Badenhop	70.2	0	2014	31	BOS	AL	70	0	0	11	0	3	.000	1	70	20	18	19	40	2.29	3.08	169	1	289	261	17	1	5	2	4	3	14	5	1	0	0	2	.268	.319	.352	.672	89	1061	679
22	Trevor Rosenthal	70.1	0	2014	24	STL	NL	72	0	0	59	2	6	.250	45	57	25	25	42	87	3.20	2.99	115	2	308	256	13	1	5	4	2	4	8	2	0	0	1	1	.223	.337	.305	.641	86	1263	800
23	Tyler Clippard	70.1	0	2014	29	WSN	NL	75	0	0	6	7	4	.636	1	47	22	17	23	82	2.18	2.75	174	5	278	250	7	1	1	1	2	2	1	4	2	0	0	0	.188	.257	.284	.541	54	1148	762
24	Kelvin Herrera	70.0	0	2014	24	KCR	AL	70	0	0	12	4	3	.571	0	54	12	11	26	59	1.41	2.69	285	0	285	252	13	0	0	3	4	0	5	6	2	0	0	1	.214	.295	.266	.561	57	1125	721
25	Cody Allen	69.2	0	2014	25	CLE	AL	76	0	0	44	6	4	.600	24	48	21	16	26	91	2.07	2.99	178	7	279	248	13	0	5	1	2	2	6	4	1	0	0	4	.194	.271	.331	.601	72	1139	732
Rk	Player	IP	GS	Year	Age	Tm	Lg	G	CG	SHO	GF	W	L	W-L%	SV	H	R	ER	BB	SO	ERA	FIP	ERA+	HR	BF	AB	2B	3B	IBB	HBP	SH	SF	GDP	SB	CS	P

Fast forward to 2014 and ironically it's the over-the-top Derek Jeter farewell tour Yankees who may most closely resemble that 1996 model.  Setup man Dellin Betances led all relief pitchers, those with zero starts, with 90 innings.  Closer David Robertson, now a free agent, had 39 saves and 64 innings.

But now three one inning relief specialists are the new model.  Let's look at those Kansas City Royals.

Rk	Pos	Name	Age	W	L	W-L%	ERA	G	GS	GF	CG	SHO	SV	IP	H	R	ER	HR	BB	IBB	SO	HBP	BK	WP	BF	ERA+	FIP	WHIP	H9	HR9	BB9	SO9	SO/W
6	CL	Greg Holland	28	1	3	.250	1.44	65	0	60	0	0	46	62.1	37	13	10	3	20	0	90	0	0	9	240	277	1.83	0.914	5.3	0.4	2.9	13.0	4.50
7	RP	Wade Davis	28	9	2	.818	1.00	71	0	11	0	0	3	72.0	38	8	8	0	23	0	109	3	0	1	279	399	1.19	0.847	4.8	0.0	2.9	13.6	4.74
8	RP	Kelvin Herrera	24	4	3	.571	1.41	70	0	12	0	0	0	70.0	54	12	11	0	26	0	59	3	0	1	285	282	2.69	1.143	6.9	0.0	3.3	7.6	2.27

As you can see, their innings in the regular season were: 62, 72, 70: 204 total innings, two fewer than Mike Marshall had all by himself in 1974 for the Dodgers in 106 games.  In fact the number of games for the three KC triplets was 206 but, of course, many of those games overlapped with two or all three appearing in the same games.

162 games multiplied by 27 outs = 4,374 outs over a full regular season.  You can probably see where I'm going.  The three headed monster accounted for only 612 (204 * 3) outs, which is 14%.

However, if they accounted for the final three innings in all their team's games that would be 33%.  So the threesome, no matter how effective, can impact less than half the games.

Since even wild card Kansas City won 89 games, what are we talking about?  Let's say that the three pitched in 70 of 90 wins.  That' 78%.  Pretty good.  But what about the other 90 games when they did not pitch?  Maybe some of those games could have been won had the manager not been waiting for his team to either take a lead after six innings or some other comparably restricted criteria.

Let's say the three pitched every other game no matter what the game situation.  That would be about 80 games, ten more than now.  But they would have no impact in half the games.  None.  They might as well be deactivated, which is something I advocate: 30 players on the active roster but only dress 25 for each game.

Dellin Betances in Baltimore
July 11, 2014
by Keith Allison
via Wikimedia Commons

The bottom line is that people who say that games can be reduced to six innings because a team has three lock down relief pitchers are mistaken.  Those pitchers are part time players.  It doesn't matter whether the model is the Lone Ranger Mariano Rivera or Betances/Robertson or Herrera/Davis/Holland.  Especially since the managers use them in a blind formula, ignoring pitch count, and simply have them pitch a particular inning, they are restricted and that restriction limits the number of games in which they pitch.

You might say that this doesn't matter because many games are not winnable and why waste these three in such games.  However, maybe many of those games that seem to be unlikely wins in the late innings could have been kept competetive had the three been used differently.

Try this case.  The first three batters in the first inning reach base and the starting pitcher cannot continue.  Which pitcher would you bring in to face the cleaup hitter with the bases loaded?  99% of the time the manager will bring in a so called long man, a pitcher who is neither good at starting or relieving.  Why not bring in your best strikeout relief pitcher?  In 2014 for Kansas City that would the 8th inning guy, Wade Davis: 13.6 SO per 9 innings.  Have him pitch to batters 4. 5. 6 and try to blow them away and then take your chances with lesser relief pitchers to get you to the 8th inning.  The "long" man is likely to allow multiple inherited runners to score and possibly some of his own because he is facing the power of the lineup.  KC might be down 5-0 after one inning and the late inning situation would justify not using the big three at all.  But using at least one of them early might keep KC in the game and make a win much more possible.

This is yet another chink in the armor of the one inning wonders who pitch less than every other game.