So Long Greg Dobbs (Or, A Lesson in Small Sample Sizes)

by David S. Cohen on Jun 22, 2010 4:02 PM EDT

As dajafi relays below, Greg Dobbs has been designated for assignment. Based on his performance this year, it wouldn't be a surprise at all if he makes it through the waiver period and finds himself playing for the Iron Pigs in 10 days. But, stranger things have happened, and he might be picked up by a major league team soon.

If he does, it certainly won't be on the basis of his 2010 performance. Instead, it will be because some GM (Ed Wade maybe?) recalls his 2008 glory days, when he was a power threat off the bench and a good fill-in at third base when Pedro Feliz needed some time to ponder how difficult it is to get on base.

Any GM who hires Dobbs based on his pinch hitting ability in 2007 or 2008 is going to be in for a powerful lesson about small sample sizes. The chart of Dobbs' career as a pinch hitter illustrates this perfectly:

	PA	HR	RBI	AVG	OBP	SLG	OPS	sOPS+
2005	26	0	5	0.250	0.269	0.375	0.644	97
2006	17	0	1	0.294	0.294	0.529	0.824	146
2007	57	2	18	0.292	0.386	0.521	0.907	173
2008	67	2	16	0.355	0.388	0.532	0.920	174
2009	60	1	4	0.167	0.250	0.241	0.491	48
2010	29	0	2	0.040	0.172	0.080	0.252	-20

sOPS+ compares Dobbs' OPS in a particular split (here, as a pinch hitter) compared to the rest of the league. 100 is average. A number over 100 indicates better than average; a number below 100 indicates worse than average.

Obviously Dobbs had a few stellar years as a pinch hitter. But, focusing on the OPS or sOPS+ columns is a big mistake. The most important column here is the PA column. In none of those years did he get more than 67 plate appearances as a pinch hitter.

To put this in perspective, 67 plate appearances is roughly how many plate appearances a regular gets in two weeks. Think of how good, mediocre, or awful a regular, any regular, can be over the course of two weeks. This is what we see with Dobbs in the chart above.

He's in the midst of a terrible stretch, and he may never recover his form. But, with small sample sizes like these, what anyone gets from him going forward is a pure crapshoot.

Tweet Comment 14 comments | 0 recs |

Do you like this story?

Comments

his .191 BAbip this season also jumps out at me.

by Wet Luzinski on Jun 22, 2010 4:17 PM EDT reply actions

BABIP may be a partial explanation, but I don’t think it explains everything; I mean, his approach at the plate was just plain awful this year. He swung at 44.2% of pitches out of the strike zone (MLB average: 28.3%) and a healthy 56.7% of pitches overall (MLB average: 45.1%), which is out of whack with what he’d done previously in his career. But he still only made contact with 79.4% of all pitches (MLB average: 81.0%) and wasn’t putting a charge into the ball at all (career worst 10.4% line drive ratio).

Or, to sum all that up: Dobbs was swinging more than he ever had, making slightly less contact, and most of the contact he made was weak.

by PhillyFriar on Jun 22, 2010 6:08 PM EDT up reply actions

i wouldn’t get too upset about the contact rate being 79.4 vs 81.0. That’s really insignificant. In my observations, he was bad at putting balls in play as I can remember a few ABs where he’d see about 8 pitches before finally missing and striking out on a 1-2 count.

by Bilzo on Jun 22, 2010 6:12 PM EDT up reply actions

My point was more that for a guy swinging at every pitch he saw, he wasn’t making the sort of Polanco-esque contact that he’d need to have any sort of success with that approach.

by PhillyFriar on Jun 22, 2010 6:17 PM EDT up reply actions

Best case scenario is that he goes down to AAA and gets his stroke back. A decent hitter who bats from the left side and can play third base (even poorly) is a nice piece to have around.

by taco pal on Jun 22, 2010 4:27 PM EDT reply actions

I agree with you there. I wouldn’t mind seeing Dobbs back.

As I have said a 1000 times, and this illustrates, the key to good benches is playing time. Dobbs got significant(for a bench player) playing time in 07 and 08 due to injuries to various guys.

If he can go down to AAA and get some playing time and hopefully his swing back then he could be a big help to us.

Looking forward to the Kevin Kolb era.
5-8-10...the day the Purdue Boilermakers basketball team won the 2011 NCAA Championship!!

by EREX21 on Jun 22, 2010 4:30 PM EDT up reply actions

Agreed. And I think it very likely that Dobbs makes it through waivers, because while he isn’t exactly making a king’s ransom, the prorated portion of $1.35 million is kind of a lot to pay a guy who’s contributed -0.7 WAR this year.

by PhillyFriar on Jun 22, 2010 6:10 PM EDT up reply actions

I’m a bit confused, this is not snark, but I’m going to cut and paste your comments so as not to mess them up.

sOPS+ compares Dobbs’ OPS in a particular split (here, as a pinch hitter) compared to the rest of the league. 100 is average. A number over 100 indicates better than average; a number below 100 indicates worse than average.

whoa….I swear on the front page his sOPS values were off by a factor of 10. Ehh…nevemind.

by Bilzo on Jun 22, 2010 6:09 PM EDT reply actions

they are…..it appears they’re truncated on the front page which is why I was confused as hell, because if 100 is average…there’s no way his 2008 campaign was a ‘17’

by Bilzo on Jun 22, 2010 6:10 PM EDT up reply actions

Yeah, it’s one of the things I don’t like about the layout on SBN front pages. For me, on both computers I’ve looked at this on, the column isn’t truncated. But, apparently, for some others it is.

by David S. Cohen on Jun 23, 2010 8:43 AM EDT up reply actions

Oh, and almost forgot to say: nice work here, David. Functions as a great example of why: (a) small sample sizes can’t be trusted; and (b) finding an “ace” pinch hitter is a dicey, if not impossible, proposition.

by PhillyFriar on Jun 22, 2010 6:12 PM EDT reply actions