Designated Hitter: Part-Time or Full-Time?

I listen the Baseball Prospectus "Effectively Wild" podcast during the week. The two hosts, Ben Lindbergh and Sam Miller of BP, cover a couple topics each day in a short podcast that is perfect as a walking-the-dog accompaniment. I don't know about you, but I often wait to listen to podcasts untill I have a batch of them cued up, then I listen back-to-back. Effectively Wild's my exception to that rule. I usually listen to it the same day I download it. It's that good.

Anyway, Sam and Ben were recently talking about the DH. You know that David Ortiz was World Series MVP. You also know that when the Red Sox were in St. Louis, they had to bench Mike Napoli and move Ortiz to first. Sam and Ben wondered whether the advantage of a full-time DH is countered by a lack of roster flexibility. The Yankees, for example, had Travis Hafner as their DH for 72 games but otherwise used the position to rehab or rest convalescing players--16 games for Alex Rodriguez, 15 for Curtis Granderson, four for Derek Jeter, etc. The lack of a full-time DH allowed the Yankees to use the position as needed given what was going on with their roster, giving them more flexibility. Actually, the Yankees, whose DHs were last in batting (.189), on base percentage (.276), AND slugging (.307) are probably not the best example of this approach. To find out which strategy's better--full-time or part-time DH--I ranked every team's DH position by OPS and then checked out how many games their primary DH played. Here's the list:

Rk		BA	OBP	SLG	OPS ▾	Primary DH, Games
1	BOS	.310	.398	.560	.958	David Ortiz, 129
2	KCR	.290	.377	.412	.789	Billy Butler, 150
3	SEA	.265	.333	.448	.781	Kendrys Morales, 122
4	DET	.302	.354	.425	.779	Victor Martinez, 139
5	TOR	.253	.347	.425	.772	Adam Lind, 61
6	LAA	.264	.341	.414	.754	Albert Pujols, 65
7	CLE	.239	.340	.414	.753	Jason Giambi, 58
8	BAL	.234	.289	.415	.704	Danny Valencia, 42
9	OAK	.230	.309	.389	.698	Seth Smith, 55
10	TEX	.245	.313	.385	.698	Lance Berkman, 65
11	TBR	.214	.307	.373	.680	Luke Scott, 63
12	CHW	.219	.290	.384	.674	Adam Dunn, 74
13	MIN	.214	.294	.341	.634	Ryan Doumit, 49
14	HOU	.198	.276	.337	.613	Chris Carter, 47
15	NYY	.189	.276	.307	.583	Travis Hafner, 72

Provided by Baseball-Reference.com: View Original Table

Generated 11/18/2013.

Wow. That's pretty stark. Four teams had a DH that played regularly. Those four teams got the most out of the position. Simple as that.

Except it's not. I looked at the prior five seasons, 2008-2012, and calculated the correlation between games played by each AL team's primary DH and the team's DH OPS. It turns out that there is no correlation at all. None (correlation coefficient = -0.01). The most DH production during those five years was from the 2011 Red Sox, who had Ortiz for 135 games. But the leading team in 2012 was the Yankees, who got fewer games out of their primary DH than any team in that period:

Rk		G	AB	R	H	2B	3B	HR	RBI	BB	SO	BA	OBP	SLG	OPS
1	Melky Mesa	1	1	0	1	0	0	0	1	0	0	1.000	1.000	1.000	2.000
2	Andruw Jones	18	38	4	11	4	0	3	10	6	9	.289	.378	.632	1.009
3	Robinson Cano	9	32	6	9	1	0	3	5	4	6	.281	.378	.594	.972
4	Derek Jeter	25	103	18	40	5	0	3	12	10	18	.388	.439	.524	.963
5	Raul Ibanez	28	92	16	26	5	0	7	22	8	18	.283	.340	.565	.905
6	Nick Swisher	12	47	8	13	3	0	3	11	6	15	.277	.370	.532	.902
7	Eric Chavez	19	54	9	15	2	0	3	7	9	10	.278	.375	.481	.856
8	Alex Rodriguez	38	150	22	46	9	0	5	20	14	38	.307	.371	.467	.837
9	Russell Martin	4	11	2	3	1	0	0	2	2	4	.273	.385	.364	.748
10	Eduardo Nunez	5	13	2	4	1	0	0	3	0	1	.308	.286	.385	.670
11	Jayson Nix	3	6	1	1	1	0	0	3	1	1	.167	.286	.333	.619
12	Mark Teixeira	4	17	3	2	0	0	2	4	0	5	.118	.118	.471	.588
13	Curtis Granderson	1	3	1	0	0	0	0	0	1	2	.000	.250	.000	.250
14	Steve Pearce	2	5	1	0	0	0	0	0	1	1	.000	.167	.000	.167
15	Casey McGehee	2	7	0	0	0	0	0	0	0	1	.000	.000	.000	.000
16	Ichiro Suzuki	1	4	0	0	0	0	0	0	0	3	.000	.000	.000	.000
	Team Total	153	583	93	171	32	0	29	100	62	132	.293	.362	.497	.860

Provided by Baseball-Reference.com: View Original Table

Generated 11/18/2013.

In 2009, the Royals had a more or less full-time DH, Mike Jacobs, who was lousy: he hit .224/.296/.374 in 106 games, as Kansas City finished last with a .655 team OPS for DHs. But the next year, the Mariners mix-and-matched (nobody more than 52 games) to a last-place .194/.269/.340 line, which in turn wasn't as bad as this year's Yankees, who had no DH appear in over half their games. 

So what's the conclusion? There really isn't one, in terms of strategy. Teams with a full-time DH sometimes do great, and sometimes do poorly. Teams that use the position as a combination gym and respite site sometimes do great, sometimes do poorly. The lack of positional flexibility is outweighed by performance when you have a good full-time DH, but isn't when you don't. It all depends on the personnel.

I suppose the next step would be to see how AL teams with a full-time DH fare in the Series compared to AL teams that don't. Another day.