I could imagine a skillful use of big data making a difference on draft day; for example, it might help a GM choose between two players at the same position by isolating key variables that are likely to predict success or failure in the pros. There are so many variables involved in successful talent evaluation that expert judgment will continue to be critical. That said, a more data-driven process might help decision makers correct for their own assumptions and biases about what attributes are most critical for an NFL player – and maybe cut down on the number of spectacular early-round busts, which are still common despite all the resources teams put into scouting and evaluation.

Once we’re using analytics to win, we obviously need the same analytics to ‘not lose’, meaning this is a wide-open field that will be developing for years to come.