There’s never been a time when football was so data-driven. Whether it’s which players to select for pro ball games or which ones to choose for college, coaches and organisations look to no one else but numbers, models and performance data to make important decisions leading to winning.

This change is not just a coaching and team thing! The growth of fantasy football leagues, as well as sophisticated stats and non gamstop sites that monitor weekly trends and performances, has gone hand‑in‑hand with the rise in football’s analytic side of the game. Coaching styles, scout procedures, and even game-time strategy are all based upon numbers.

Why Football Teams Started Believing in Analytics

Data was made truly available when technology brought both player and game information within reach. GPS-based tracking systems and video analysis software, combined with statistical models, then entered the picture. With the aid of these, teams gained access to data of an order that, up until then, could only have been dreamed of: efficiency of route running, trends of defensive coverage, etc.

It was not until a statistical modelling-based analytical department was added to the current team of football coaches that it became a runaway success in the NFL. Eventually, colleges started creating similar systems, all with the same intent: finding areas that only the naked eye could not reliably uncover, the proper path for team success.

Video analysis gave coaches the ability to break down individual tendencies of opponents. The player tracking included acceleration, separation and workload points. Modelling would help teams build possible statistics of their players, taking individual success of various team systems and match situations into account.

How Analytics Has Changed Fourth Down Decisions

The calls from the college and pro football sidelines have consistently been of a conservative nature for many years. After teams got to fourth down, they were able to punt or kick a field goal. The thinking was that you must return the ball, or face the music, regardless of how the gamble paid off.

The analytics have since systematically undermined all of the above. Expectancy Points models have shown that the number of points you must give up on a fourth-down punt in specific field-position situations is greater than what you might gain on a successful conversion. Win Probability models will now show that an aggressive fourth-down call at short yardage often gives a team a better chance of winning the game despite the risk of giving the opponent the ball.

It has taken coaches who come into the league with a more analytics-based approach to implement the findings from these models, though. A staff that bases its decision on analytics would compare the points they’d concede on a punt with the odds of converting prior to calling on a punt.

The algorithms that have been created now take into account possession of the ball, the margin of score, and time, rather than just the first or second factor alone, and have been enhanced regarding their time management. In this day and age, you can be certain someone on the sidelines, or on TV, will be equipped with exactly the same data as a coach, and they will scream if he makes a bad fourth-down decision. Or a bad decision when they are up against a simplistic model.

The Role Of Analytics In Player Recruitment

NFL front offices are increasingly relying on the combination of both scouting and statistical analysis. When this is done, it is the inclusion of contextual efficiency, positional value, and efficiency ratings that gives analytical departments their deeper understanding of potential recruits.

In the realm of player recruitment, the main function of analytics is to find hidden talent. It helps discover players that have not been playing well (so their raw numbers are not turning out to be as great as they should), like a very good receiver whose numbers seem low, but are very good when adjusted for the situation.

Moreover, analytics has also shown that each position value is actually in direct correlation with the odds of winning. Some positions have more impact on the win than others. This discovery has become a tool for teams to figure out the allocation of draft capital and contract money.

Analytics isn’t the end-all, be-all of scouting, though. Character evaluation, medical evaluations, coach and locker room culture will always need the traditional testing measures.

The Reasons For Continued Coach Scepticism

As one can expect, not all coaches have embraced this statistical revolution; there exists a certain mentality in certain camps where counting can’t resolve everything. Furthermore, there are countless non-countable factors that do not enter analytical models because they can’t be measured; i.e.:

• Emotional momentum
• The effects of the crowd
• Leadership under duress
• Individual player traits

Many who have participated in football throughout their lives believe that it is something that can be applied in real ways. It seems the ideal solution is to use analysis to assist decisions, rather than have it take over.

Conclusion

The influence that statistical analysis has had on how decisions are made in football is permanent. NFL and college football coaches have come to rely on it in every area of their job: coaching, drafting, managing and everything in between. Instinct and experience still have their place as well, but models and metrics continue to evolve.