What is the key to rushing the passer?
In the NFL, pass rushing is becoming exceedingly more and more valuable as the league shifts away from the run-heavy offenses of the past into the modern day pass-happy systems of today. With a heightened emphasis on the quarterback position, it only seems logical that defenses would prioritize rushing the passer in order to affect the quarterback’s ability to make quality throws. What’s the key to success as an NFL pass rusher?
The NFL salary data defends this importance placed upon rushing the passer. As of 2023, NFL pass rushers were paid as the third highest position in the league, first among defensive positions. On average, each NFL pass rusher made $3.95 million a year, behind only quarterbacks and offensive tackles.
The values for all salaries by position can be seen in Figure 1.
When looking at the top ten highest paid players of each position, pass rushers once again rank favorably, as the position finished 4th among the 11 positions. The top ten pass rushers made an average of $19.2 million, behind quarterbacks, tackles, and defensive tackles, a position which on occasion can also be considered pass rushers.
In this study, NFL pass rushers from the 2015–2021 Drafts were analyzed in order to determine which of their statistical factors most highly correlated to NFL success during these seven NFL Drafts. These factors, which included numerous measurable metrics, athletic test scores, and collegiate statistics, were analyzed in order to create a statistical profile of each player.
In order to properly analyze the collected data, collegiate statistical data was obtained from PFF, while measurables and athletic test scores were gathered from Pro Football Reference and RAS, an athletic test score metric created by Kent Lee Platte. RAS compares the entire athletic profile of one player to the standard athletic profile of that position. The metric is on a scale from 0–10, where 0 is the worst statistical athlete ever at the position and 10 is the best.
Firstly, it is important to define success. What is success? How is success measured? Sacks are the statistic most relied on to view a pass rusher’s success. This, however, can be very misleading. A player can have a lot of impact creating pressure on the quarterback, while not getting the necessary sack numbers to be considered “effective”. Total pressures is a much better statistic, however, injuries must be taken into account as well.
The company PFF provided the total pressures per year, as well as, how many games each player had played thus far into their career. The pressures were added and divided amongst the games played. This allowed for players who had been injured not to be negatively affected. This metric (PPG) took into account that some players do not see the field as often, diminishing their impact as a pass rusher and avoided the problem of small sample sizes if done on a per pass rush snap basis.
For the NFL, this was the best metric available in this study to measure success at rushing the passer. The pressures per pass rush snap statistic will come into play later.
It was found that a player from the pool of 176 averaged 1.56 PPG. For this study, the threshold for success will be set at 2.0 PPG played. Of the 176 players, 52 were able to reach this mark, roughly 33% of drafted players.
With this definition of success in mind, the analysis was under way. What available statistics had the highest correlation to NFL PPG? With so many variables in the equation, it was apparent no correlation would be too strong. If it were that simple, the draft would be a sure bet.
Many of the NFL decision makers have preconceived notions as to what metrics or attributes best correlate to success in the NFL. Jacksonville Jaguars general manager, Trent Baalke, holds the belief that arm length and wingspan are some of the biggest factors in determining a player’s success in the NFL. In a 2023 interview, Baalke stated, “It’s a big man’s game, in our opinion. We’re looking for length, I think length over time has proven to be successful in this league.”
Baalke’s draft history backs up this idea. In Baalke’s time as GM for the Jaguars and San Francisco 49ers, he has never drafted a player who has under 33-inch arms.
Last year, Baalke selected Georgia defensive end Travon Walker. Walker, widely regarded as a premier athlete, had rare arm length at over 35 inches.
The idea behind the precedent is that the longer one’s arms, the further the player can keep an opposing player away from them, controlling the direction and pace of the engagement.
Statistics, however, suggest a different story. When looking at NFL PPG, arm length has a weak negative correlation, indicating that as PPG increases, arm length is either a non-factor or even slightly decreases.
There are many other beliefs that NFL decision makers hold. Many teams have size or athletic thresholds that players must meet in order to be a possible selection for that respective team. Another example is height with quarterbacks. In much of NFL Draft history, height has been a common threshold among NFL teams. Recently, however, with a shift to a more athletic mold of QB, that height minimum is not as common as the league has now drafted Kyler Murray, Tua Tagovailoa, Baker Mayfield, and recent draftee Bryce Young.
Whether all of these thresholds are substantiated by statistics is to be determined, however, for edge rushers, arm length does not correlate to success. That begs the question, what does in fact correlate to success.
In order to calculate that, correlation tests were run, looking at what athletic metric revealed the highest correlation with NFL pressures per game. That answer was Lee Platte’s RAS. The correlation between RAS and NFL pressures per game, while not strong, had the strongest correlation by a good margin at 0.24. Out of all of the athletic metrics, it was a high RAS that gave players the best chance of success at the NFL level.
The scatterplot for the correlation is shown in Figure 2.
Out of all players shown in Figure 2, just three players below a 5.0 RAS met the 2.0 PPG threshold. Players in this study below the 5.0 RAS statistically had just a 17.6% chance of meeting the threshold. Out of the same players, just 13 players of 64, below 7.5 RAS, met the mark or roughly 20%.
To contrast, of the players above a 7.5 RAS, 42% had a 2.0 PPG or higher. This is a telling sign for all of those pondering the validity of RAS. While not strong, it does correlate to increased NFL pressures per game.
Following determining which athletic metric had the highest correlation, it was time to discover which collegiate performance statistic could do the same.
Out of all the collegiate statistics analyzed, it was another statistic created for the purpose of this study that best correlated to success. The value pressures per pass rush snap was defined as the new Pass Rush Efficiency (PRE). It was this PRE metric that had the highest correlation with NFL PPG.
While pressures per game were used for the NFL statistic, it isn’t surprising that pressures per snap is the best metric in college. In the college data, statistics were only derived from a drafted players’ last season in college. Due to the vast majority of the drafted players being starters, PRE makes sense as the highest correlated metric to pressures per game in the NFL.
The correlation value between the two statistics came out to 0.60, a decently strong correlation. The scatterplot of the two variables is shown in Figure 3.
Of the players in this study, the average PRE was 0.1374 or 13.74%. This indicates that on average, a player in this study pressured the quarterback on 13.7% of their pass rush snaps.
In total, 100 players scored below the 0.15 metric. Of those 100, just 16 players would go on to average over 2.0 PPG in the NFL as seen in Figure 3. Alternatively, 63 players scored above the 0.15 mark for PRE. Of those, 34 scored above the PPG benchmark or about 54%.
Between all the athletic and collegiate statistics, it was PRE that was most closely correlated to that of NFL PPG. If only total pressures were evaluated, the correlation value would have been that of 0.169. It was imperative to view pressures on a per snap basis, because the pass rush snap differences between players in college were so vast.
Keeping the high correlations of RAS and PRE in mind, the next goal was to look at the two in unison. The average RAS was about 7.6 and the average PRE was that of 0.1374 as mentioned previously.
To conclude, the study looked at those players who scored above a benchmark RAS of 8.0 and a PRE of 0.15.
With those filters set, 29 players drafted between 2015 and 2021 remained. On average about 76% scored above the set 2.0 NFL PPG, a very favorable percentage of successful players. If teams are looking for an elite pass rusher, these filters set one of the best models for success at the NFL level for pass rushers. The 29 edge rushers averaged a very impressive 2.82 pressures per game.
Those 29 players are shown below in Figure 4.
In this data analysis of edge rushers, it was RAS and PRE that were most highly correlated to success at the NFL level. There will certainly always be players that score poorly in one or both of these metrics that end up being successful athletes. However, there are no other metrics widely available that will be better correlators for success.
Some other findings that were interesting was the correlation between draft age and NFL success. While not something the player can change, draft age had a higher correlation to success than most every other metric. The correlation value between the two variables was -0.46, a moderate, negative correlation. This indicates that as age decreases, the NFL pressures per game rises. While not as strong a correlation as PRE, it was notably higher than most other variables and something teams should take note of.
One last noteworthy discovery was draft value for these edge rushers. A person might expect the more successful edge rushers to be taken in the first round, and the data strongly backs up this theory. The average player gets drafted between the 3rd and 4th round. However, when looking at the top 50 players of PPG in the NFL, the average round the players were drafted in came out at 2.13. The NFL historically has drafted edge rushers early and values them highly. This study shows they have had every right to do so.
The NFL Draft is one of the most televised events in sports, garnering well over 10 million watchers annually. This three day event, in some instances, can make or break a team’s future. Sometimes a team is really just a player or two away from winning. With such uncertainty surrounding the draft, this study gave statistical insight into what metrics best account for success in the NFL at one of the league’s premier positions.
There are dozens of statistics that can be evaluated and it is unlikely the draft will ever become strictly data oriented. This analysis, however, showed that there are metrics which correlate with NFL success. Of those in the data set, RAS and PRE provided the best correlations to PPG in the NFL. With this study’s results, other fans, data analysts, and teams can improve upon their knowledge of the NFL Draft and the edge rusher position.
There is the possibility for further examination as future draft’s edge rusher prospects can be studied to understand which players have the best chance for success. In addition, similar tests can be run for other positions. Of the other positions, wide receiver and cornerback best fit into a quantitative analysis, such as this one, due to the athletic and statistical nature of their positions.
It will be exciting to see what similar tests are to follow as the NFL attempts to slowly adapt to a statistical approach to the game. As is the saying, “adapt or die.” What other myths are out there to be tested? One day, maybe the NFL does end up cracking the code as the sport of baseball has done.