By: Ryan Dunker and Tommy Schilder
Introduction
Among the most pivotal moments in a football game is the two point conversion. Whether it’s necessary to tie the game, or a bold move to take the lead, it’s hard to find a two yard play that is more consequential than the two point conversion.
Despite the two point conversion being a relatively rare play, sports fans have seen some teams risk it all to make it big. In Super Bowl LI, the Patriots made history by coming back from a 28-3 deficit. Aside from scoring touchdowns, they were able to tie the game at 28 and later win the game by successfully converting two two-point conversions.
It might seem rather simple to just get 2 yards, right? Well, over the 2017-2020 season, NFL teams successfully converted a two point attempt only 49.2% of the time. With odds of success that low, if you’re going to go for two, there’s clearly a lot of pressure to make the right play call. So what should you do, run or pass?
Data Breakdown
In an effort to answer the question, we collected and analyzed two point conversion play data from STATHEAD (a branch of Pro Football Reference) from the 2017 through 2020 seasons.
From this data, we are classifying a play based on its result, not necessarily what was called in the huddle. For example, a designed pass that results in the quarterback running will be classified as a rush. In total, there were 461 attempted two point conversions by all 32 teams during these four seasons. When it comes to analyzing all these attempts, here’s what we found:
[*Note: of the 461 plays, 9 of these attempts (from the 2017 season) were disregarded towards run or pass data due to lack of description provided by the data. These plays were marked as successful or unsuccessful, however we could not discern whether they were run or pass plays.]
TWO POINT ATTEMPT SUCCESS RATES: 2017-2020 NFL SEASONS
Figure (1)
After looking at this graph of play percentage over the last four seasons, we can see that pass plays have occurred nearly three times as much as rush plays. Even with a slight dip in league-wide pass percentage, there is still a strong affinity for pass plays over rush plays when going for two.
Even if rushing plays are not usually the play of choice, it is also important to note that the percentage of rush plays have steadily been increasing each year. With passes simultaneously decreasing, this could be an indicator that we may see a more balanced spread of rushes and passes across the league in future NFL seasons.
Figure (2)
Ironically, rush plays have been more successful in each of the past four seasons despite its much lower preference among teams. Even this past season, at its closest distance to pass success rate, rushes were successful at 51.28% compared to 46.73% for passes. It should also be noted, however, that rush success has been on a continual decreasing trend, while pass success has improved slightly from 2017 to 2020.
With a league success rate of 49.2% on two-point plays these past four seasons, it’s hard to ignore the fact rushes have been 57.14% successful, nearly a solid 8 points above the league average.
Despite the fact that rush success from 2017-2020 was nearly 12% higher than pass success, it is important to note the steady decrease in rush success and increase in pass success each season. These trends could suggest a possible reversal of distributions in seasons to come, especially considering the notion that teams' rush play percentages are steadily on the rise (Figure 1).
Figure (3)
In the chart above, we have 64 points to represent both the passing and rushing components of the 32 NFL teams (from 2017 through 2020). In doing so, we sought to measure the overall success rate versus play rate for each individual team when it comes to rushing and passing.
At first glance, one can see a stark difference in success rates for rushing plays when a team calls them less than 25% of the time. These polar opposites are largely due to the overall lack of rushing play calls by some teams.
For example, the Bengals attempted seven total two point conversions from 2017-2020. Of those seven, only one was a rush attempt. This one rush attempt was successful, making the Bengals technically have a success rate of 100% when running for a two point conversion. So in a sense, success rate for teams with a low rushing rate will be either heavily inflated or deflated due to the lack of significant attempts.
Besides this initial discovery, one can see that teams with a passing rate greater than 0.75 are heavily clustered in the 20-60% success range. Alternatively, teams with less than a 0.75 passing rate have a greater spread, as the success rate varies slightly more.
Although not concrete, this does suggest that teams with relatively similar play call rates will be just as successful if not more successful than teams with higher passing rates. In other words, teams that have more balance between rushes or passes, if not more preference for rush plays, appear to have a higher chance of being successful on any given 2 point attempt.
4th Quarter Investigation
While we are primarily interested in determining the best play call for a two point attempt at any given point in a game, one can’t help but question whether two point attempts in the fourth quarter have a different distribution of play calls and play successes. This is because many times in the fourth quarter, these conversions have live-or-die implications, so there could be a different play calling strategy with this element in mind.
Figure (4)
This graph resembles a similar trend to that of play frequencies by year at any point in the game. This is primarily because most two point attempts come in the fourth quarter. Ultimately, we can conclude that passes are clearly more preferred, even in game-deciding two-point plays.
Figure (5)
Just like the frequency chart for 4th quarter attempts, this chart also closely resembles play success rates for any point in the game. Rushing success appears to be slightly higher in the 4th quarter versus at any point, while pass success rates appear to be the same as at any point in the game. Although pass rates are not currently as successful as rushes in the 4th quarter, it seems that offenses are making an effort to convert more on passes late in games, while defenses have gotten better at stopping the run when the game is on the line. It is worth monitoring if rush attempts will in fact continue to decrease for 4th quarter attempts, even though they have clearly been more successful.
Conclusion
The two point conversion has always been a risky play in football, and even though the success rate of a two-point conversion has only been 49.2% over the past four seasons, the expected value of going for two is as mathematically sound as going for an extra point. With this in mind, teams should be looking to make a play call that maximizes the probability of success, especially in end-of-game situations. In these scenarios, what type of play should the coach call, a rush or pass?
Given our analysis above, there was substantial evidence to determine that rushing plays are more successful than passing plays. In fact, the overall success rate for rushes is 12% higher than the success rate for passes over the past four seasons.
Despite this, teams are still passing the ball more often than they are running. Although there is no concrete explanation for this, reasons for more favoritism towards pass plays could be due to the high confidence in receivers to catch goal-line fades, or perhaps a lack in confidence of offensive linemen to win the leverage battle against stacked defensive lines. At the end of the day, the data does suggest that many pass-oriented teams should reconsider their play calling habits for a two point conversion.
In our analysis, we also found it notable that the rush success rate has been steadily decreasing over the past four seasons, while the pass success rate has been slowly increasing. This trend could very well hold for years to come, particularly as the league continues to be filled with creative, bright-minded offensive coordinators.
Overall, we were able to uncover interesting information regarding team success when it comes to rushing and passing as well as play calling. This said, we conclude that, when taking every team in the league into consideration, a rushing play would be a better play to call on a two point conversion. With this conclusion in mind, it is important to take into account some rush success inflation (like with the Bengals) as well as the decline in rush success. With the data currently favoring a rush play, it will certainly be worth checking back in future seasons to see if rush plays still remain more successful when going for two.
Sources: stathead.com