Courtesy of the Seattle backfield, two of the most wrongheaded fantasy football charts I’ve ever seen.
First, from the Best Ball ADP App:
And from the Dynasty ADP App:
As you can see, Thomas Rawls is valued multiple rounds ahead of C.J. Prosise in both formats. That’s backwards. Prosise should be valued ahead of Rawls in all formats.
Let’s Start With The Obvious…
Prosise was a third round pick in the 2016 NFL draft. Rawls went undrafted in the 2015 NFL draft. It is true that Rawls has some of what you’re looking for in an undrafted free agent success, including the fact that he was a college workhorse. But let’s not forget the obvious: Generally speaking, you want the RB with the higher draft pedigree. Generally speaking, when presented with two young RBs, you would expect the RB with the higher draft pedigree to get more opportunity. In fact, draft position helps explain RB opportunity for the first two years of a RB’s career. In the linked article it actually states that draft position is more explanatory in a RB’s second year than in the first, so there’s not much reason to believe Rawls is out of the UDFA woods just because he had a good rookie year. Note that is true even when you include prior year carries and yards per carry, usually the factors people point to as being in Rawls’ favor.
Another obvious point: RB receptions are valuable, and you would, generally speaking, prefer the RB who is a better receiver. Rawls only has 20 receptions in his combined collegiate and NFL careers, compared to 430 rushing attempts. Prosise had 29 receptions just last season and actually played wide receiver in the two prior seasons. There is a non-zero chance that Rawls is the better receiver, but that’s the most generous thing I can honestly say on the issue. There is no actual reason to believe that he is, and significant reason to believe that Prosise is better by a significant margin.
One counter to the above point might be that Seattle historically hasn’t used their RBs much as receivers, so Prosise’s receiving value actually isn’t very valuable. But I think the obvious counter to that is their primary RBs have been Rawls and Marshawn Lynch, neither of whom are highly regarded receivers. Don’t get me wrong, they’re not necessarily bad receivers. But it’s probably not a coincidence that both have consistently had large rushing workloads with relatively small receiving workloads. Despite being 34 years old and only getting 26 rushing attempts, Fred Jackson had 32 receptions last season. The fact that Jackson saw a Lynch level of receptions while only getting a fraction of the carries and never playing more than half of the snaps in a week suggests to me that the lack of RB usage in the receiving game actually is a personnel issue. For instance, if Jackson had seen a Lynchian workload, meaning a dramatic increase in rushing attempts and snaps, it only seems reasonable to believe he would have also seen a significant increase in receptions.
In fact this may be a situation where people are just (probably unintentionally) finding a way to spin a problem with Rawls into a problem with Prosise. If nothing else, it wouldn’t make sense for Seattle to draft a RB who is widely-regarded as a good receiver and not use him as such, and they’ve actually been giving him reps at WR. Seattle had 79 RB targets last season; I wouldn’t be surprised if Prosise’s presence leads to an increase in targets, which he then dominates.
Then there’s the issue that even Rawls’ rushing performance was unsustainable. In games where Rawls saw at least 10 carries, he was on a pace that would have added up to 336 carries over the season. That would have not only led the league, but would have been more than Lynch had in any season of his career. He also averaged 5.6 yards per carry. Jamaal Charles is the most efficient rusher in the history of the modern NFL, and he has only averaged 5.5 yards per carry for his career. Note that the 75th percentile mark league-wide for a RB1 is about 4.55 yards per carry. So even if Rawls is a very talented rusher who gets a workhorse load, he probably won’t be putting up the efficiency or the volume that he did last season.
Arian Foster May Have Gotten Lucky
A lot of people think or hope that Rawls will emerge as the next Arian Foster, the next big UDFA RB success. There are two problems with that. The first is obvious, and discussed above – Foster was a great receiver, Rawls is not. But the second is less obvious, because it’s the part of the Arian Foster story that people tend to conveniently forget.
After a solid rookie season in 2009, Foster had a breakout sophomore season in 2010, racking up almost 150 yards from scrimmage per game. But it’s what happened in the middle that may have been the most important. In the 2010 NFL draft, the Texans selected RB Ben Tate in the second round. Then Tate got injured in the preseason, which resulted in him missing the entire season.
This brings us to the million dollar question: Would Foster have broken out if Tate hadn’t gotten injured? Considering the draft position issues discussed above, and the fact that Tate still managed to total 1,040 yards from scrimmage as a sophomore even after Foster broke out, I think the answer may very well be no.
This may seem like a pointless digression based on a counterfactual. After all, Rawls isn’t Foster, Prosise isn’t Tate, and who knows what would have happened if Tate hadn’t gotten injured? But I think it’s an important point. A lack of competition may be a necessary condition for a Foster-like breakout for Rawls, and if it is, that’s a condition that isn’t present currently. Another good example is C.J. Anderson, who only got a shot after the more highly drafted Montee Ball and Ronnie Hillman got injured in 2014.
But maybe I’m being overly narrow. Maybe I should expand it to later round draft picks as well. There have been eight RBs drafted between the fifth and seventh rounds of the NFL Draft who saw at least 150 touches as a rookie. Three of those represent somewhat favorable outcomes for Rawls. There’s Alfred Morris, but he’s kind of a bad comp because he was a 16 game starter and Washington didn’t add a meaningful RB in the offseason.1 There’s Andre Ellington, whose team also didn’t add a meaningful RB in the offseason. It should be noted that they did, last offseason with David Johnson, and that RB has since supplanted Ellington. Tim Hightower had a better sophomore season, though that was on the back of 63 receptions, and he was largely irrelevant after that. The Cardinals actually did add Beanie Wells in the preceding offseason.
One of those RBs in particular jumps out at me as an example of the risk Rawls poses: Zac Stacy. Stacy had a strong rookie season, though not as strong as Rawls. But it was strong enough that drafters didn’t hesitate to spend a premium pick on Stacy, even after the Rams drafted Tre Mason in the third round. Stacy went on to total just 445 yards from scrimmage and get traded to the Jets the following offseason.
Let’s broaden it even more to all RBs who had solid usage and efficiency as rookies. From 2005 to 2014, there were 31 RBs with at least 150 touches who managed at least 4.0 yards per carry as rookies. Of those 31 RBs, in only five cases did a team add a more highly drafted drafted RB the next year. I’ve included a table with those RBs and their rookie (year N) and sophomore (year N+1) yards per game below:
While all but Isaiah Crowell had fewer yards per game as a sophomore, I wouldn’t put too much stock into this table at all. None of these RBs are great comps for Rawls, and it is a very small sample.
But it does change my mind about one thing. Some Rawls backers have argued that the fact that Seattle drafted three RBs shouldn’t be perceived as having anything to do with Rawls directly. In the past, I’ve agreed with that position despite being a Prosise backer. But I see now that it has actually been quite rare for teams with a successful rookie RB to draft even a single RB higher than their successful rookie. So the fact that Seattle drafted three such RBs should not be dismissed out of hand. Shawn Siegele thinks the Seattle backfield is so wide open that Alex Collins could actually be the back to own. I believe Collins represents a bigger risk to Rawls than Prosise because Prosise has the draft position advantage over Collins and Prosise is the only one of the three to distinguish themselves as a receiver.
A Weighted Coin Flip
An overly simplistic way to look at which RB should be ranked higher would be to ask who is more likely to finish with the most points. A better way to look at the issue though, would be to look at expected value. After all, one may have more value than the other if they end up being the starter. This brings us back to the receiving issue. Given their respective receiving histories, it seems fair to say that Prosise is the one that is more likely to have a big receiving season. Since it is increasingly the case that RBs who catch a lot of passes dominate the position, Prosise would seem to have more upside. Which means that not only does Rawls need to be likely to outscore Prosise to be the better pick, but so much more likely that it’s enough to counterbalance Prosise’s higher upside.
Maybe I’m Wrong…
…But I don’t have to be right. Prosise isn’t valued ahead of Rawls, and that doesn’t look likely to change anytime soon. If Prosise should be valued ahead of Rawls, then I’m getting a huge discount. If not, then I’m paying less for a lesser back.
Contrast that with the opposite position, which if you act on means spending a premium draft pick on a RB with a far worse draft pedigree and minimal receiving production. That position has much less room for error, and thus requires a lot more confidence. You might even say it requires overconfidence. As you can see, Rawls’ ADP has barely declined in either format since the Seahawks drafted Prosise and Collins. Maybe Rawls’ pre-draft ADP just happened to almost perfectly reflect the possibility that they would draft a Prosise-quality RB, but it seems to me that drafter overconfidence in their pre-draft valuation of Rawls is a far simpler explanation.
We can generalize this issue beyond Rawls and Prosise, and Charles Kleinheksel did that just a few years ago when he assigned RBs to cohorts based on their ADPs. Prosise and Rawls are what Charles would call, “small gap RBs,” RBs where the difference in their ADP is less than the league average for teammates, which was approximately 100 picks at the time.2 What Charles found at the time was that the secondary back (which would be Prosise in this case) was only outscored by the primary back (Rawls) by about 2.5 FPPG. Adding in 2014 data, the secondary back has outscored the primary back 23 out of 56 times, 41 percent of the time. Going anecdotal for 2015, this cohort produced Devonta Freeman, David Johnson, Giovani Bernard, and Duke Johnson.
Don’t draft Rawls. Draft Prosise, who is almost certainly the better value, and probably the better pick outright.