Building Sustainable NFL Contenders

A look at correlation between salary cap allocation by position and three-year winning percentage

Introduction and Methodology

There have been numerous studies that have looked into how NFL teams should optimize spending by position in order to maximize win percentage or expected wins. However, these analyses tend to look at the issue using a one-year lens. In the NFL, it is important to win sustainably and over a longer period of time. As such, I sought to see if there was any correlation between cap allocation by position and three-year winning percentage.

To do so, I pulled positional spending data from OverTheCap between 2013-20. I then used the historical standings found on ProFootballReference to calculate forward three-year win percentage for each team dating back to 2013. So for example, a team’s 2020 forward three-year win percentage would capture their wins and losses from the 2020, 2021, and 2022 seasons. I chose this time period as it felt like an appropriate window to analyze the true quality of a team, while also aligning with the three-year value metric that teams like to use to analyze contracts. While injuries or luck can certainly impact a single season, teams that are bad for three straight years are likely to see changes implemented in their Front Office while those experiencing three years of success will likely be considered one of the better teams in the league. I then calculated the correlation between cap spend at each position (as a percentage of cap) and three-year win percentage to see if it could tell us anything about building sustainable contenders in the NFL. Allocation strategies of course change year-to-year, but I thought it would be useful to evaluate if certain ones helped setup long-term success.

As a next step, I sought to analyze the 25 largest contracts signed between 2013 and 2020 (based on guarantees and excluding rookie contracts) at each of those five positions to see if that data provided us with additional insights on the correlation between positional spend and winning. As part of this analysis, I looked at standard contract details as well the player’s PFF grade in the year before signing their deal as well as in their final year under that contract with their original team. In instances where the player was traded or released mid-season, I tried to use my best judgement whether to use his grade from that season or the one before. Usually this meant considering current season performance when the transaction occurred after mid-October. For deals that have not yet expired, I used 2022 performance as a fill-in for the end-of-contract grade. I also looked at how many games the player missed and the dead money taken on at the end of the player’s tenure with the team that signed them. Deals that will likely take on dead money in the future due to existing void years were considered to have zero dead money for the purposes of this exercise, given that teams could theoretically sign these players to an extension and avoid or push off the charge. The goal of this was to see if I could identify any trends in either performance or contract structure that could help further explain the correlation I had found.

Analysis and Key Findings

In analyzing the data, I found statistically significant correlation between spend at Tight End (0.21), Safety (0.17), and Cornerback (-0.24) at a 99% confidence interval and at Interior Defensive Line (-0.16) and Edge Rusher (0.15) at a 95% confidence interval. To be clear, none of these correlations are especially strong, but they do clue us into some commonalities between teams that have found ways to win sustainably. However, there was no correlation - either positive or negative - between winning and spend at the majority of positions. This tells us that there are many ways to build a successful team.

Compared to other positions, Tight End contracts are very reasonable. On average, the APY (average salary per year) as a percentage of the cap at signing was 5.2% for the contracts analyzed. Overall, the average allocation of cap to the TE position during this time period was 5.3%. Not all these signings worked out, but given the reasonable guarantees, teams were able to get out of the deals without incurring too much dead money - in just two instances was the total over $10m. An illustration of how reasonable TE contracts is that the San Francisco 49ers George Kittle’s $15m APY topped his position, a deal signed one year after the Browns inked Wide Receiver Jarvis Landry to a deal with an APY of $15.1m. Landry also received $7m more in guarantees. While Landry is no doubt a good player, he is far from being at the very top of his position like Kittle. Furthermore, Kittle and Landry had comparable numbers as receivers in the year prior to receiving their contracts, but Kittle also added tremendous value as a run blocker. This helps show that great TEs are a sound investment as teams can sign them for a similar price to a good WR, but with the added benefit of contribution to the run game. Indeed, many of the most successful teams over the past decade have had top TEs - such as the Patriots with Rob Gronkowski, the Chiefs with Travis Kelce, the 49ers with Kittle, and the Eagles with Zach Ertz and Dallas Goedert.

Teams spend slightly more on Safeties than on Tight Ends, but on average, this group is the lowest paid on the defense. Average cap allocation to safeties was just 7.2% across the time period I examined. While few would argue that Safeties are more important to a team than Cornerbacks, a look at their contracts shows why investing in Safeties could be a sound financial decision. The average APY as a percentage of cap for the Safeties analyzed was 6.4% compared to 8.3% for Cornerbacks. Yet, at the time of signing, these Safeties had an average PFF Grade of 79.6, compared to 74.1 for the Cornerbacks. A similar phenomenon emerges here as with TEs and WRs, where teams can get a better quality Safety than Cornerback while spending less on the Safety. Given their contributions in both run support and in pass coverage, Safety might be a slightly undervalued position in the NFL. Again, a look at some of the top teams over the past decades shows that several made relatively large investments at Safety such as the Patriots with Devin McCourty, the Seahawks with Kam Chancellor and Earl Thomas, the Chiefs with Tyrann Matthieu, and the Eagles with Malcolm Jenkins. Even in the 2000s, teams like the Ravens with Ed Reed, Buccaneers and Broncos with John Lynch, Steelers with Troy Polamalu, and the Patriots with Rodney Harrison reaped the benefits of spending on premier Safeties.

Edge Rushers receive large contracts and teams devote significant resources to the position. The APYs of the top-25 contracts given out to Edge Rushers accounted for an average of 9.5% of the cap at signing. Teams, on average, dedicated 11.8% of the cap to the position. The position is integral to the strength of a defense considering its importance in both rushing the passer and stopping the outside run. Given that, it makes sense that these players tend to be the highest paid on the defense. That being said, while the contracts are large, teams do seem to be rewarding the right players. The Edge Rushers I analyzed had an average PFF grade of 81.5 and median of 83.9 in the season prior to receiving their extension or free agent contract. While performance does tend to decline towards the end of the contract, just as it does at other positions, we have seen players like Joey Bosa, Myles Garrett, and Demarcus Lawrence play high-level football throughout the duration of their big contracts. That’s not to say that every big-money deal has worked out, but that by and large teams are doing a good job identifying the right players at the position to pay handsomely. Given their importance, NFL teams will likely have to continue to devote significant resources to Edge Rushers - although the sheer influx of talent over the past several years might make it easier to get talented players at the position at a lower price point. Still, it is difficult to think of any successful teams over the recent years that did not have good Edge Rushers on their rosters.

Cornerbacks, on the other hand, seem to be a bad investment despite being a premium position in the NFL. As mentioned before, the APY of the top CBs accounted for an average of 8.3% of the cap at signing. Teams devote an average of 10.5% of the cap to the position. However, teams seem to be overpaying in Free Agency for the talent they are getting. The average PFF grade at signing for the CBs I analyzed was 74.1 and the median was 73.8. Those numbers dropped to 65.0 and 63.8 respectively by the end of the contract. While the drop-off in production is comparable to other positions, Cornerbacks grade out far worse than players at other positions by the end of their contracts. Given that deals are mostly backloaded, this means that teams are often dedicating significant chunks of the cap to players no longer playing at a level consummate with their pay. Athleticism is essential to the CB position, so it makes sense that play drops off as players approach their late 20s and early 30s - especially with the highly-paid boundary CBs. Furthermore, scheme changes can further cause performance to decline if the player is not a fit. Cornerback is still an important position, but teams should be cautious in giving out large free agent contracts. The Patriots signing of Stephon Gilmore in 2017 and Seahawks re-signing of Richard Sherman in 2014 are examples of big deals that paid dividends, but Washington’s signing of Josh Norman in 2016 is an example of a deal that did not work out. Overall, the data shows that teams should be cautious about trying to find a CB1 in Free Agency, and try and build the top of the depth chart in the Draft.

Lastly, a look at the top contracts for Interior Defensive Linemen might help tell us why spend on the position is negatively correlated with winning. The APY for the top IDLs accounted for an average of 9.0% of the cap at signing. Teams on average dedicated 10.5% of the cap to IDLs. That figure is in line with spending on CBs, and not much lower than on WRs (11.4%) and Edge Rushers (11.8%). In today’s game, however, IDL is not widely considered to be a premium position. Yet, the pay is more in line with positions like WR, Edge, and CB than it is less premium positions like Linebacker, Safety, and Tight End. The top IDL contracts tend to go to players with pass rushing prowess, yet the contracts tend to be in line with those given to Edge Rushers. The phenomenon that exists with being able to get better TEs than WRs and Safeties than CBs for the same money or less does not exist when comparing IDLs and Edge Rushers. Additionally, IDL performance tends to decline a little more than it does at Edge Rusher. The IDLs analyzed saw their PFF grades decline by an average of 11.1 compared to just 8.3 for Edge Rushers. IDLs ended their contracts with an average PFF Grade of 69.3 compared to 73.2 for Edge Rushers. Furthermore, while the data is not overwhelming, there is evidence that pressure from the Edge is more valuable than it is from the interior. As such, teams might be better off paying for Edge Rushers to generate pass rush instead of pass-rushing IDLs given that the contracts are comparable. Teams can instead focus their efforts on signing IDLs that are effective against the run, given that those players tend to be lower cost and fill an important role on the defense, and spend that money elsewhere.

Concluding Thoughts

Ultimately, there are numerous ways to build a good football team. For example, there are plenty of examples of teams that have spent big at Cornerback and Interior Defensive Line and been successful. Few, for example, would argue that the Los Angeles Rams would have won the Super Bowl without having Aaron Donald and Jalen Ramsey. Ultimately, teams all operate under the same salary cap, and need to try and assemble the best roster possible under that hard constraint. With that in mind, the commonality between these findings is that they help identify some ways that teams can use trends in positional value to get similar traits and production for less money. If teams can sign a better player at TE than at WR, and pay the same money or less, it helps them assemble a better football team while not sacrificing the quality of pass catching weapons. If investing in a premier Safety allows a team to spend less on Cornerbacks - contracts that we have seen generally age poorly - it could free up resources to spend elsewhere. And if teams have to spend the same to get pass rush prowess from Edge Rushers and Interior Defensive Linemen, it makes sense to focus resources where that investment will generate more impact and bring in a higher quality player. The margin of victory in the NFL is often small, and while the correlations I found were not strong, even the smallest advantages in roster building can help teams win more football games.

Teams ultimately need to be willing to adapt to win sustainably. The NFL is a copycat league, and once teams see certain strategies work, they tend to adopt them. It is almost certain that were I to conduct this analysis 10 years from now, the findings would be different. It is possible, for example, that teams placing an increased value on Tight Ends will reduce the gap in quality between a comparably priced Wide Receiver, thus erasing that competitive advantage. It is unlikely we’ll ever see strong correlation between spend at a position and winning - there are far too many other variables at play. But analytics and data can continue to complement good scouting to help teams build the best possible roster under the constraints of the salary cap.

I am happy to share the Excel workbook I used with anyone who requests it. Please don’t hesitate to reach out via email at gauravv@mit.edu.