How a Madden Score Can Improve NFL Talent Identification

by Fusion Sport
 | 27th July, 2021

By Tyler Lindon and Alex Campanella

Sports media and fans are obsessed with player stats. It’s not surprising the last full NFL Combine in 2020 saw a 119% increase in viewership. TV personalities and amateur analysts pay rapt attention to how fast each player can run, how high they can jump, and how nimbly they can weave through cones.

While these simple metrics, like a 4.2 40 or 46” vertical, entertain arm-chair GMs and as they cast their draft order predictions, they do not paint a complete picture of player potential.

Experienced NFL scouts know that finding the best fit requires collecting player data set from multiple sources, viewing it in context of his game film, and carefully considering the team’s playing style and roster needs. While their jobs have become increasingly data-driven, it is still very much a people business and the instincts and experience of veteran team personnel is irreplaceable.

That being said, we believe there is an opportunity for NFL teams to improve talent identification and development by bridging the gap between objective and subjective data. It is possible to create a holistic rating system for players similar to the “Madden Score” many football coaches, scouts, and performance staff have told us they wished they had.

The Case for Change

Researchers from the University of Louisville found that posting strong Combine numbers do not mean a player is more likely to thrive in the NFL. The authors noted that while the event offers some useful indicators, “Combine exercises simply measure athletic skill and not actual football-playing ability.” If they did, Tom Brady would not be the owner of seven Super Bowl rings.

In Brady’s case, he was able to more than compensate for his comparative lack of physical prowess through his superior technical, tactical, and psychological capabilities (see Fergus Connolly’s book Game Changer for more on this model).

The authors of the University of Louisville study also referenced the comprehensive player profile created by Bill Walsh and his staff in the heyday of the San Francisco 49ers, which sought to create a more complete picture of each draft prospect by incorporating insights gleaned from game film and what coaches had to say about their character.

We’ve seen the success of this approach in other sports more recently, such as the person-first, player-second methodology the Kansas City Royals pursued when assembling the squad that went on to win the 2015 World Series.

 

Finding the Telling Metrics

While the Combine offers a useful snapshot of physical ability, teams can widen their lens by analyzing historical data on each player and zeroing in on telling metrics that relate to the kind of player they are looking for. This would be comprised of subjective and objective elements and give equal weight to physical, psychological, technical, and tactical acumen.

Rugby provides an example of this approach, where some coaches use the amount of time spent on the ground after a tackle as an indicator of the player’s effort level in practice. Experienced football personnel, armed with historical data, can identify similar metrics that don’t merely illustrate a player’s physical capacity but also hint at how committed and hard-working he is.

Once this standardized assessment is developed, the results can be collected in an athlete management system like Smartabase and layered over the top of historical performance data collected in practices and games. This combination would increase the amount of context scouts had for every prospect and get us much closer to a more holistic Madden Score type of system.

Our work with NFL teams and other elite sports organizations from around the world makes us uniquely qualified to solve complex human performance challenges such as this. If you’d like to explore how Fusion Sport can help your team improve how it uses data to identify and develop talent, please contact us.

 

If You Enjoyed This Article, You May Also Like…