Elite running backs are a precious commodity in dynasty fantasy football. Dave Caban outlines a simple way to maximize your chances of finding tomorrow’s stars.

Earlier this month I used regression tree analysis to better understand wide receiver prospect profiles. Regression trees help us to understand the mixture of attributes that tend to drive NFL performance and provide a visual way to understand how these attributes interact. I built a regression tree in that article that could be used to estimate the points per game that a WR prospect would score in his first three NFL seasons.

Since completing it, I’ve been working on a similar analysis for running backs. I’ve yet to arrive at a model I’m ready to publish, but from using a variety of training sets and workshopping different regression trees, I have identified a couple of factors that tend to be useful when considering RB prospect profiles.^[1]Regression trees built using RB prospect data tend to overfit to the limited data available. In particular, there are two factors, that when paired together, provide very competitive hit rates in identifying backs with the potential to be elite fantasy players.

Key Statistics

A majority of the regression trees I looked at included