Drafting Rookie Running Backs: A Primer

Note: This is a free Dynasty MetriX related article to give you a sense of the kind of information you’ll have access to with a premium membership to Dynasty Football Fan. Dynasty MetriX is a rating framework for rookies based on the revolutionary database of statistics that we’ve developed. Our predictive model helps you gain an edge on the competition. Our premium membership gives you access to Dynasty MetriX related analysis, our unique statistically based rankings, our 28-page 2017 Dynasty Rookie Draft Guide and our 260 page 2017 Prospect Fact Sheets. Join the Dynasty MetriX revolution now and dominate your league! 

Drafting rookie running backs is an emotional endeavor. It’s extremely hard to get quality, productive RBs in rookie drafts if you’re not in the top half of the first round as the obvious (or seemingly obvious) choices are off the board in the first four or five picks. Annually, we reach for RBs that seem to be in prime situations even though their metrics don’t indicate a level of athleticism or college production that would translate to NFL success. 2017 is no different as we’re seeing players like Kareem Hunt and Samaje Perine go in the first round in rookie ADP. Are Hunt or Perine worth a first round pick? How about Williams (both Joe and Jamaal) or Mack in the second round? This is where Dynasty MetriX (DMX) can help. Here is the current RB ADP from MFL with corresponding DMX scores:

Leonard Fournette           0.85             1.7
Christian McCaffrey           1.78             3.2
Joe Mixon           1.37             4.3
Dalvin Cook           1.12             5.6
Samaje Perine           0.41           12.2
Kareem Hunt           0.69           12.8
Alvin Kamara           0.77           13.3
D’Onta Foreman           1.24           22.9
Marlon Mack           0.53           24.9
Joe Williams           0.27           25.8
Jamaal Williams           0.27           26.8
James Conner           0.34           35.8
Jeremy McNichols           0.78           39.3
Wayne Gallman         (0.02)           43.2
Aaron Jones           0.65           49.6
Elijah Hood         (0.30)           55.8
Donnel Pumphrey           0.44           57.5
Brian Hill           0.62           64.0
Elijah McGuire         (0.07)           65.4
Tarik Cohen           0.73           65.6
T.J. Logan         (0.02)           68.5
Corey Clement         (0.35)           69.0
Matt Breida         (0.14)           71.8
Devante Mays         (0.67)           72.4
De’Angelo Henderson         (0.47)           72.6
Matt Dayes         (0.59)           72.7
Christopher Carson         (0.19)           79.0

Here’s a quick refresher on how DMX is developed and how it can be used, which will provide you some context. Once you read through those articles come back and take a second look at this list. The first thing you notice is that Leonard Fournette is going tops with a 0.85 DMX score. While I have detailed why this should concern you, Given his injury his final year in college I can get over picking him first given what we’ve seen when he is on the field. The next three RBs have elite DMX scores and are completely justified to be taken in the top 5 or 6 picks as is being done. Then something interesting happens. Perine, Hunt and Alvin Kamara jump D’Onta Foreman in ADP, but have much lower DMX scores. Foreman’s DMX score is elite, yet being behind Lamar Miller he is getting no respect. Meanwhile, Perine and Hunt have been hyped up as the next best things in Washington and Kansas City, despite the competition in place on both teams.

Then you move below Foreman and you have Mack, the two Williams’ and James Conner with relatively low DMX scores compared to Jeremy McNichols. Same kind of situation, Mack and the Williams’ (sounds like the name of a jazz band) have quicker opportunity to get on the field than McNichols, so they are being drafted higher.

Finally, you see guys like Aaron Jones (who is in Green Bay with Jamaal Williams), Brian Hill and Tarik Cohen being drafted late, among the RBs with negative DMX scores (never draft those guys). Aaron Jones has a higher DMX score than Jamaal Williams, but is being drafted 23 spots below him. WTF? What gives? Let me give you some points to consider and then provide some facts that may sway your opinion on some of these lower DMX RBs being drafted so high.

  • There is no other way to put it, D’Onta Foreman and Jeremy McNichols are being under-drafted given their DMX scores. If you are into quality, long-term prospects, getting these guys at these prices should be a priority.
  • Historically, based on my prediction model, only 4% of negative DMX score RBs hit, which is defined as performing within the top 20% of all RBs since 2001 within their first 5 years in the league. Guys like Wayne Gallman and Elijah Hood should be ignored.
  • While all positive DMX scores provide a reasonable chance for a prospect to hit, as we’ve defined it above, relative value is an important consideration. Why take Perine at 12 when you can take Foreman at 23? Why take Hunt at 13 when you can take McNichols at 39?
  • Another issue to consider is other positions. David Njoku and Evan Engram are being drafted around 12/13 as are Perine and Hunt, but their DMX scores are 1.59 and 1.69 respectively. Patrick Mahomes is going mid-second with a DMX score of 1.52. Zay Jones is going mid-2nd with a DMX score of 1.49. These are all elite DMX scores with estimated 70% hit rates associated with them. Why take Perine or Hunt who’s estimated hit rates are between 15%-34%?

The reason is emotional as much as it is practical. Perine and Hunt are being hyped and if you need an RB (and we all need a RB!), without DMX to help guide you, the choice seems easy. DMX tells a different story though and here are some facts to guide your decisions.

First let’s consider whether to draft Perine, Hunt, Mack, the Williams’ or Connor within the first two rounds of the draft. The place I’d like to start is with looking at their speed score. Hunt, Perine, Jamaal Williams and Connor have below average speed scores (below zero in the DMX context) while Mack and Joe Williams’ speed scores are positive. Let’s take a look at speed score relative to 5 year production:

This first graph is a scatter plot with speed score on the vertical axis and 5 year fantasy point totals across the horizontal axis. To the right of the red vertical line represents where the top 20% reside.  The black oval to the right shows 4 prospects out of 298 prospects with negative speed scores since 2001 who have had elite production in their first 5 years. To be clear, that’s 1.34% of the negative speed score population. Those four prospects were LeSean McCoy, Ricky Williams, Brian Westbrook and Arian Foster. All of these four, save for Foster, were drafted higher than Hunt, Perine, Jamaal Williams and Conner. Note, you can see that above zero on the vertical axis there are 14 prospects with elite production and positive speed scores, compared to the 4 in the oval below zero speed score.

  • This conclusion alone would prevent me from choosing Perine or Hunt with a first round pick. Both have negative speed scores. Are you willing to use a first round pick on a guy who has only a 1.34% chance at elite production?
  • Jamaal Williams and Conner are a bit better value in the second or third round range, though I’d still prefer McNichols or even Mack or Joe Williams to these two, since they have positive speed scores and Jamaal Williams and Conner do not.

The oval to the left shows the vast majority of prospects with negative speed scores rarely produce any fantasy points at all. Certainly there is a grouping between those two ovals that produce small to moderate fantasy production, but do you want to waste a top two round pick on small to moderate production as the likely outcome?

The second graph here is another scatter plot that compares total DMX score to 5 year fantasy point totals:

When considering this scatter plot, keep in mind that Perine, Hunt, Mack, both Williams’ and Connor are all between 0.00 and 0.69 DMX scores. This scatter plot is set up in a similar way. The black oval to the right of the red line shows the prospects with 0.00 to 0.69 DMX scores that have hit since 2001. There are 10 of them in there, even though it looks like 4. They are Chris Brown, Ahmad Bradshaw, Alfred Morris, Brandon Jacobs, Domanick Davis, Joique Bell and Chester Taylor, Kevan Barlow, Tim Hightower, and Michael Turner. My point here? None of these guys were elite. Turner had a couple of elite years later in his career before regressing again. Morris, Bradshaw and Jacobs all had there moments but over there career were not worth top tier investments. Perine, Hunt, Mack, both Williams’ and Connor are in that same boat. None of them profile as elite talents but they will have a good year here and there given the rise of an opportunity and their development in the NFL. The vast majority of 0.00 to 0.69 DMX score RBs fall to the left of the red line in the left black oval, with little or no production.


I love how deep this RB draft is, despite what you may perceive as negativity above. When you look at the facts regarding these mid-tier RBs it’s clear that many of them are being drafted over and above their potential due to off-season hype. Below is a chart on where I value these RBs in question given what the DMX prediction model tells me. Given where some of these guys are currently being drafted, it will make them undraftable in my mind. Perine and Hunt are not worth a first round pick. Jamaal Williams is not worth a second and Connor is not worth a third. I’d rather pick high quality at other positions, or other higher quality but low perception RBs in those spots.

Join the DMX revolution!

Round 1 Round 2 Round 3 Round 4 Round 5 Undrafted
Christian McCaffrey D’Onta Foreman Kareem Hunt Joe Williams James Conner Wayne Gallman
Joe Mixon Jeremy McNichols Marlon Mack Samaje Perine Aaron Jones Elijah Hood
Dalvin Cook Alvin Kamara Jamaal Williams Brian Hill Donnel Pumphrey
Leonard Fournette


This site uses Akismet to reduce spam. Learn how your comment data is processed.