clock menu more-arrow no yes mobile

Filed under:

Very minor correlation between NFL heavy personnel usage and offensive effectiveness in 2023

A deep dive into NFL trends

Buffalo Bills v Miami Dolphins Photo by Megan Briggs/Getty Images

NFL offenses don’t remain static philosophically.

What was effective and popular a decade or two ago isn’t likely to be what’s effective and popular in today’s league. The constant morphing of philosophy to counter what defenses are doing in regards to scheme and personnel is one of the reasons why being an informed football fan is so important: you can’t just say “this is how an offense should run” and spit out the same answer for 30 years straight without being very wrong at some point.

With the Shanahan and McVay offensive systems heavily utilized throughout the league and with hot offensive coordinators like Ben Johnson getting head coaching interviews (and then, apparently turning down offers), it would be prudent to once again find out “what’s working” in the NFL.

One topic that comes up when discussing what “modern NFL offensive philosophy” actually means is the trend towards heavier offensive personnel with more frequency. While Sean McVay still meaningfully favors 11 personnel (one tight end, one running back, three wide receivers), teams like the Detroit Lions and San Francisco 49ers have begun to utilize “heavy” personnel (defined here as 12, 13, 21, or 22 personnel) a significant percentage of the time and even (in rare cases) the majority of the time. Make no mistake, it’s still an 11-majority league overall, but there is a growing trend toward heavier personnel.

But is it working?

Is there actually a correlation between heavier personnel usage and offensive effectiveness?

If, for the purposes of this piece, we define “offensive effectiveness” as “offensive EPA/play,” there exists a very slight correlation between heavier personnel usage rate and success on the offensive side of the ball.

In terms of correlation coefficients, a mark of “0” represents zero correlation. A result of “1” is a perfect positive correlation (as one data element increases, the second data element increases at the exact same rate) and a result of “-1” is a perfect negative correlation (as one data element increases, the other data elements decreases at the exact same rate).

The correlation that exists between “heavy” (12+13+21+22) personnel and offensive EPA/play is .155 (rounded to .16 if you’d prefer hundredths of a point). This represents a very slight positive correlation.

Ultimately, a correlation this slight would be viewed by many data scientists as being too small to be conclusive.

It’s something we should be keeping our eye on as offenses continue to evolve in the NFL, but any idea that heavier personnel has somehow become so obviously “the answer” to modern defenses in the league is probably early. The offensive personnel that a team has will still best dictate how it can attack defenses. Sean McVay’s Rams had the lowest “heavy” personnel usage in the entire league at a paltry 4.9% total. The reason McVay can stay in 11 on offense is very similar to why the Buffalo Bills can so frequently stay in nickel: they have the people to do it.

For McVay, he specifically targets receivers who can perform blocking duties in line at a high level like Robert Woods and Puka Nacua, allowing him to run concepts in the rushing attack that other teams may need to swap personnel to get heavier in order to have function properly. For the Bills, the presence of nickel cornerback Taron Johnson has allowed them to play the run with lighter personnel on defense due to Johnson’s ability to play downhill as good if not better than the majority of third linebackers in the NFL.

We should remain vigilant about where the modern NFL offense is trending. There is a constant ebb and flow of defenses learning to counter offenses, which then need to adjust to counter defenses. The rising prevalence of Vic Fangio-style two-high defensive systems has caused some teams to adjust with a more potent rushing attack and heavier personnel, and we should monitor this for any potential increases trend data. But for now, the correlation coefficient says it’s not an obvious “press this button and you’re more likely to have success” option but rather, something to monitor.


...and that’s the way the cookie crumbles. I’m Bruce Nolan with Buffalo Rumblings. You can find me on Twitter and Instagram @BruceExclusive and look for new episodes of “The Bruce Exclusive” every Thursday on the Buffalo Rumblings podcast network!