#1 By: RotoViz, April 12th, 2018 10:00
Originally published at: http://rotoviz.com/2018/04/three-down-backs-in-the-2018-rb-class-introducing-the-backfield-dominator-rating/
In the past we’ve measured RB contributions in the receiving game in one of two ways. Either we measure their raw counting stats — i.e., the total number of receptions and receiving yards (which does have predictive value) — or we measure their receiving contribution as a percentage of team receiving production, via the College…
#2 By: dkpsports, April 13th, 2018 04:58
I like the article, and it is good idea, but I can't help but think the sample seems to be really small, especially in those RBs coming from schools that don't throw to their RBs. Sometimes they don't throw to their RBs all that much because they are all not that good at catching the ball. Would it be helpful to look at how much production they had compared to historical comps at the same school? Most coaching staffs/systems are fairly stable so may be a helpful indicator.
It would help if we could see some correlation with NFL performance, even if it is only anecdotal?
#3 By: Blair Andrews, April 13th, 2018 10:41
These are fair points. It's certainly possible there are schools that don't throw to their RBs because they don't have any good receiving backs on the roster. But I tend to think it's more often scheme or QB related. Previous work has shown that running QBs do not target RBs as often, for example. Regarding comparing previous players from the same school, I'm not sure of the real value there. Even if college coaching staffs are stable (and it's not clear they are significantly more stable than NFL staffs), college rosters are not. Every team is completely different from the same school's team just four years prior, for all intents and purposes.
Ronald Jones fares well in my analysis, considering he only caught 14 passes. Just a few years earlier, though, Buck Allen caught 41 passes at USC. However, that was a completely different team with different personnel (and in this case, a different head coach), so trying to draw a conclusion from this sort of comparison seems tenuous. I'll keep that in mind though, and give it some more thought.
We have enough examples of RBs who didn't catch a lot of passes in college but went on to be great receiving backs in the NFL that it's worth asking why they didn't catch passes in college -- BDR is one way of trying to answer that. Jamaal Charles, for instance, never had even a 20-catch season at Texas. The most receptions LaDainian Tomlinson ever had at TCU was 16. Obviously I'm cherry picking the best examples, but the point is that raw college receiving stats are often deceiving. I'm hoping BDR can correct that to some degree.
I'm in the process of testing it against NFL performance and determining the best way to use the metric -- more to come soon.
#4 By: dkpsports, April 15th, 2018 13:42
I just looked. With regard to Ladanian Tomlinson and jammal Charles, neither were the primary receiving backs for their schools in their senior year. Both had yardage percentages below 40 percent it seems.
#5 By: Blair Andrews, April 16th, 2018 20:04
I don’t have BDR data on Tomlinson but that doesn’t surprise me—he was actually quite bad at receiving in college. I do know TCU did not throw to RBs often while he was there. I think he had half their RB receptions (even if his share of yards was much lower—he only had about 4 yds/rec—like I said, bad).
Charles accounted for about 47% of Texas’ RB receiving yards his senior year, according to my numbers. (His BDR was about 64%—above average, but hardly notable.) That’s not a lot, but for someone who caught only 17 passes it’s significant. Guice caught 18 passes but only 15% of LSU’s yards. So whereas Guice was really only used as a two-down back for LSU, Charles actually had a significant (but not dominant) role in the passing game (despite the presence of Chris Ogbonnaya who was strictly a third-down back).
My initial tests reveal BDR is probably most useful for late-round picks. RBs with the sort of draft capital Tomlinson had rarely fail, and so BDR adds little.