The N.F.L. playoffs, which begin Saturday afternoon, will probably bring high-stakes moments around a simple decision: After a touchdown, should your team kick for one point, or go for two?
The 2-point try — where teams get to run one play to score from the 2-yard line — was initially a hit with coaches when it was introduced to the N.F.L. in 1994, but its popularity soon faded. In recent seasons that trend has reversed, with 2-point attempts becoming more common thanks to rule changes and to the growing role of analytics.
For fans, this trend has brought a welcome dose of strategy to what was once an automatic decision to kick. (Week 17’s game between Detroit and Dallas, which hinged on not one or two but three consecutive 2-point attempts in the game’s final seconds, typified both the strategic and emotional factors at play.)
Here’s a primer on what’s changed and what’s behind the decisions you’ll see in the weeks ahead.
Why are attempts going up?
The short answer: a rule change. In 2015, the N.F.L. moved the spot for the extra-point kick from the 2-yard line back 13 yards, to the 15, making a successful kick slightly less probable (it’s now successful 95 percent of the time, on average, instead of 99 percent). That decrease was enough to make going for two more appealing.
How often are 2-point attempts successful?
About 47.5 percent of the time since 2015 — almost exactly half that of the extra-point conversion rate.
Interesting! Was that a coincidence or by design?
It’s hard to say exactly how much math was used in the N.F.L.’s decision to move the ball to the 15 for teams who choose to kick. (Much closer and the math says they should kick; much farther away, going for two is clearly the better choice.) But the league stated it wanted to make it more entertaining for fans.
Is there a straightforward, math-based strategy for when to go for two?
For most situations, there is consensus (among statisticians!) for when to go for it. However, in some situations, different assumptions can lead to different recommendations.
If a team scores a touchdown that leaves it 5 points down, the math is very clear: Going for two is the right choice. A successful conversion would put the team within a field goal of tying the game. A failed conversion or a successful kicked extra point would leave a deficit of 4 or 5 points.
If a team scores a touchdown that leaves it 4 points down, however, the math is a bit trickier. A comprehensive analysis by FiveThirtyEight recommended going for two, especially late in the game, but a separate analysis by a University of Wisconsin-Madison professor, Laura Albert, concluded it’s best to kick the extra point. Even on similar questions, slightly different assumptions or data can lead to different answers.
So are teams mostly following the math?
I would never describe N.F.L. teams as “mostly following the math,” but their decisions are increasingly going in the same direction as the math.
Teams are more conservative than they probably should be, but they’re also now much more likely to go for two when the math heavily favors doing so.
Take that down-by-5-points example above. From 2010 to 2013, teams went for two in those situations 25 percent of the time. But in the last three years, they’ve more than doubled that rate, going for it 56 percent of the time.
Going for two when down 9 is a more complicated decision (FiveThirtyEight declared it a tossup). From 2010 to 2013, teams didn’t attempt this even once. But in recent seasons it’s becoming more accepted, with coaches facing this situation choosing to go for two about 32 percent of the time since 2020.
If you want to see a longer list of recommendations for different situations (and what teams have actually done), we’ve included a table farther down.
Why don’t coaches just outsource the decision to math? Seems like it would make their lives easier!
Math is not and should not be the only thing coaches rely on. A team may have a play (like the one Dan Campbell of the Lions used against Dallas) that it loves from practice and that it thinks has a high chance of success. Or, alternatively, weather or a specific personnel matchup could mean a play is less likely to succeed than a calculation might otherwise suggest.
But it’s also true that N.F.L. coaches have typically refrained from going for two until the end of the game, when the specific situation forces them to make a tough decision.
Still, the trends say math is winning. Coaches have started to go for two earlier in the game.
Are some teams better at 2-point conversions?
The data seems to say so.
Determining the difference between skill and luck (or statistical noise) on 2-point conversions is a challenge because they remain relatively rare. In the 2023 season, there were over 1,300 touchdowns but only 127 2-point attempts.
But there appears to be a connection, even if it’s obvious: Teams with better offenses are more likely to succeed than teams with worse offenses.
If you slice things even thinner, the top 10 percent of projected offenses per game converted at a whopping 59 percent. Projected offenses in the lower third come in just above 40 percent. A big difference!
Does that mean bad offenses should almost never go for two?
Not quite. On Dec. 31, the lowly New York Giants scored against the favored Los Angeles Rams to cut the lead to one with over three minutes left and, after a Rams penalty put the ball on the 1, decided to go for two instead of the tie and a chance of overtime. Let’s imagine that the offensively challenged Giants had a 40 percent of converting the attempt. You might think overtime would be more appealing — until you consider that underdogs of 3 or more points have won overtime only 41 percent of the time since overtime rules were revised in 2017.
In fact, the math would suggest it’s smart for underdogs to play for the win because bigger swings in a shorter time span (more variance, in stat-speak) increase the chances of an upset. Coaches have embraced this, being more risk-friendly when they are underdogs.
So the coach must make a big difference.
Yes. This year Jonathan Gannon, the first-year coach of the Arizona Cardinals, went for two on 36 percent of all touchdowns. By comparison, Kyle Shanahan of the 49ers did not attempt a 2-point conversion on any of his team’s league-leading 61 touchdowns.
This difference is partly because of differences in teams’ strengths, records and game situations, but not entirely. As coach of the Philadelphia Eagles, Doug Pederson went for two 37 percent of the time in 2020 during his team’s 4-11-1 season, a rate that remains a season record.
But three years earlier, when his team was much better (13-3), Pederson went for it 17 percent of the time — significantly lower, but still the second-highest rate in the N.F.L. that year.
By contrast, Campbell’s Lions went for two after 12 percent of his team’s touchdowns this season.
What about when a penalty moves the line of scrimmage, as happened with the Lions?
The math certainly does not shine favorably on the Lions’ decision to go for it from the 7-yard line after a penalty moved them back from the 2. (Only about a third of attempts have been successful from that mark, the same as for fourth-and-goal attempts from that distance.)
On the other hand, thanks to a 2019 rule change, teams can take advantage of penalties committed against them during the touchdown to start the ensuing 2-point attempt from the one-yard line, rather than from the 2. That additional yard makes a difference: About 59 percent of attempts from there have been successful, and teams should go for it even in a situation where 2 points are not essential.
The playoffs are here. Should we expect more or less aggressive behavior for this decision?
It’s hard to say. Since 2015, the percentage of 2-point attempts after a touchdown has been 12 percent in the playoffs and 9 percent in the regular season.
This could mean coaches are more aggressive, but the sample size for playoff games is small, so it could just be statistical noise. It’s worth noting, though, that playoff games aren’t especially closer contests than regular-season games. That doesn’t mean you can’t shout at your TV louder during the playoffs, though.
I sometimes see coaches with a big laminated play sheet on the sidelines. Are they using that to determine when to go for two?
Most likely not. That real estate is generally devoted to plays and personnel packages. And a majority of the time coaches will know if they want to go for two based on their own set of rules, without having to consult a piece of paper.
Teams do often have a “game management coach” who sits in the booth with information related to 2-point conversions; fourth downs; timeouts; when to challenge a referee’s decision; and so on. More than half of N.F.L. teams have a dedicated analytics department that would probably help produce such a guide, though coaches may create them on their own.
Now what about that table?
You don’t need to laminate it, but there have been many 2-point conversion charts created over the years that used an analytical framework.
Below we have highlighted the recommendations of FiveThirtyEight’s Benjamin Morris, which are generally in agreement with the other charts’ recommendations. Next to those recommendations are what teams have actually been choosing to do in recent seasons. Although coaches are definitely more conservative than what the general math says one should do, in recent seasons they are at least trending toward the numbers.
