Metrics Scouting: Small Running Backs

Over the last few months, I’ve been working on a way to more efficiently scout future NFL players using workout numbers and college production. I know I don’t have nearly enough time to scout every running back prospect who declares, but Microsoft Excel has no such time restraints. The below rankings list is based on my last 2 months of work fine tuning this method for smaller running backs.  Like any scouting method, this is not perfect, some players are far too high, and some are far too low. While reading this, I recommend you don’t think of this in black and white terms. For example, just because Player X is rated than Player Y doesn’t mean that Player Y isn’t as good as Player X. The grades the computer comes up with merely suggest that a player has NFL talent, not guarantee it. Somewhere around the low-80s is where players go from potential starters to role players.

The Physical Grade is average of the players Speed Score, Explosion Score, weight adjusted Agility Score, and Strength Score. They are weighted in this order: speed, agility, explosion, and strength. The Production Grade is the average of a players grade for yards/game, touchdowns/game, and receptions/game, with the receptions/game number worth less than the other two. These grades are compared to the numbers for historical elite small running backs. The final grade is the average of the Physical/Production grades, with the Physical weighted more than the Production.

The system can be fooled, no question about it. If a player doesn’t perform a drill at the combine or their pro day, I have no choice but to assume average numbers. This is the reason players like C.J. Spiller and Jamaal Charles aren’t ranked higher. Poor production in their final year is another blind spot of this system, and the reason for Maurice Jones-Drew’s low ranking. You also can’t take players prior to 2000 on this list very seriously, as combine data before that point is sketchy, at best. Marshall Faulk, Barry Sanders, and Brian Westbrook are ranked, but all of them have 2+ drills missing.

You could argue that Demarco Murray and Darren McFadden don’t belong here, as I set a rough weight limit of being under 215 pounds to be included. Murray’s listed NFL weight is now 215 and McFadden’s is 218. They’ll be on my ‘big’ running backs list as well.

Every change I’ve done has been to increase the correlation of my rankings to weighted-Career Approximate Value. The r-squared of my rankings to wCAV is 0.3537. The players NFL draft position r-squared to wCAV is 0.3406. Since I don’t know how NFL teams ranked these running backs, where they were drafted is a crude way of approximating that. It could easily be argued that wCAV isn’t a good way to determine how talented a player was, and I agree. If a player never gets an opportunity, how can we judge how talented they are? This system thinks that Bobby Rainey is a top 20 talent, but how is he going to get a chance being stuck behind Ray Rice and Bernard Pierce? I promise I’m not crazy, guys.

I encourage you to hit me up on twitter or the comments section and let me know what you think. I’m open to any and all ideas that will help me improve what I do. Without further ado, my small running back rankings dating back to 2000: