In today’s pass-happy league, the No. 1 concern for defenses has been affecting the play of the quarterback. Allowing time from a clean pocket to make a throw spells doom for defenses.

We have seen the massive contracts given to high-end pass rushers who make game-changing plays. Nick Bosa, Chris Jones, Josh Allen, Brian Burns, and T.J. Watt all find themselves in the Top 25 in average annual contract value where that list is primarily made up of quarterbacks and a few receivers. 

To that point, we set out to project the next crop of sack artists using some of our metrics. The big pain with projecting pass rush performance, though, is that context plays a huge role. An interior player who primarily plays on early downs has a much harder time than a situational pass rusher coming off the edge. 

We created Expected Pressure Percentage and Pressure Percentage Plus-Minus a few years ago for this reason. The former uses game situation, player alignment, and quarterback drop type to estimate how likely a player is to generate pressure on each play, and the latter is how much the defender performed above or below that expectation.

In order to better predict year-over-year sack production, we looked into four stats to see how predictive each of these are towards the following year’s sack output.

– the previous year sack percentage 

– pressure percentage 

– expected pressure percentage 

– pressure percentage plus-minus of pass rushers 

 Once we dove into what has predictive value, we looked at some interesting candidates for positive or negative sack regression in 2024.

Methodology

Sack percentage, pressure percentage, expected pressure percentage, and pressure percentage plus-minus were aggregated for every player in their previous season going back to 2019. 

These four metrics were then used as features in linear regression models to predict the sack percentage of the player’s next season. Players were only considered if they had 100 pass rushes in each season.

The significance of each variable and adjusted R² values of each model were analyzed to see which metrics have the best year-over-year predictive value. The adjusted R² value is a 0-to-1 measure of how well the features explain the variability of the dependent variable, in this case current year sack percentage, and is adjusted based on how many features there are in the model.

Results

For starters, all of these are statistically significant, meaning that an increase or decrease in one of these features meaningfully impacts the following year’s sack percentage. 

Also, we used sack percentage instead of actual sacks because we need to factor how many pass rushes a player had in a given season. Playing time can be affected by a variety of factors, so we are keeping it simple by focusing on per-play performance.

Sack percentage comes in with the second lowest adjusted R² value when predicting next season’s sack percentage. Considering that sacks are more output than process when it comes to good pass rush, this makes sense. Sacks come at such a low sample that having one more or fewer can drastically impact this percentage, especially year-over-year.

Expected pressure percentage shows a slightly higher adjusted R² value than pressure percentage and are the two with the best explanation of variability in next season’s sack percentage. This is the process driven argument, as getting more pressure on a quarterback shows better predictive value than your sack percentage from the previous year.

Pressure percentage plus-minus comes in at the bottom. This metric strips out the opportunities to get a sack and looks solely at the skill portion. Volume is a big part in getting sacks and situational factors are very impactful, so the low explanation in variability here (low adjusted R²) makes sense.

The best model combines expected pressure percentage with pressure percentage plus-minus, coming in with an adjusted R² of 0.32. Factoring in the skill component with the expected measure makes the model better explain variability, and it does so better than just using pressure percentage because it weighs the situational factors more strongly.

With the latter model leading the way, let’s use that to predict some sack percentages for 2024.

2024 Outlook

Decline in Wattage

T.J. Watt once again led the league in sacks for the 3rd time in 2023 with 19. However, his unimpressive expected pressure rate of 10.3% would suggest he is in for a drop in sack percentage by 1.8 percentage points in 2024. This would be a massive drop off by Watt’s standards, putting him roughly at 10 sacks this upcoming season based  if he had roughly 500 pass rushes. Leading the league in sacks three times is very impressive, but sustaining that production over time is a very tall task.

Trading Trey?

Trey Hendrickson walked back his trade request with the Bengals by expressing his desire to win a Super Bowl for Cincinnati. Hendrickson, who had a career high 17.5 sacks with the Bengals last season, is a candidate due for a drop in sack percentage by 2.2 percentage points. His expected pressure rate put him in the middle of the pack last season at 10.1%. Looking at a sell high opportunity, the return on Hendrickson might be enough to warrant a trade.

The Youthful Cameron Jordan

Cam Jordan may be past his prime at age 34, but he can still affect the quarterback at a high rate. His 10.3% expected pressure percentage last year came only with a 0.5% sack percentage. According to the model, this should result in a bump to 11.5% in his 14th season in the league. Can Jordan turn back the clock? Maybe not, but he should have higher production than his 2.5 sacks a season ago.

A More Hungry Lion

Aidan Hutchinson is projected to take another step forward and increase his sack percentage to 2.2% this year, good for another half-sack. Hutchinson took a leap last season, ranking 8th in pressure percentage among those with 30 pressures and 12th in sacks. Another step forward would be huge for a Lions team coming oh so close to getting to the Super Bowl last season.

Sack Title Contenders

Looking at the predicted sack percentages for next year and using the pass rushes a player had last year, Maxx Crosby (13) and Aidan Hutchinson (13) are our top contenders to take home the sack crown, with Danielle Hunter, T.J. Watt, Micah Parsons, Nick Bosa, and Khalil Mack just behind. Josh Uche, who had only 214 pass rushes last season, is projected to have a higher sack percentage this year at 2.2%. Playing with Matthew Judon on the other side, Uche has a chance to accumulate a high sack total given more opportunities.

Conclusion

Using advanced pressure metrics gives us more of a sense of predicted output going forward. Sacks are the output, but their predictive power is minimal. Looking at the process of getting a sack leads us to seeing who may be a diamond in the rough poised for a breakout season or someone who might be overvalued. After all, this is arguably the second-most important position in football, so getting these players right is critical to success.