New NFL overtime rules: Big questions on coin flip strategy

We got our first overtime frame of the 2025 NFL season in Week 2, when the New York Giants and Dallas Cowboys ended regulation tied at 37 after three scoring plays in the final minute. The Giants won the coin toss in overtime and chose to have the second possession. Neither team scored on their first drive -- Dallas ultimately won on a 46-yard field goal with just 4 seconds left in overtime -- but it brought up a question: Should NFL coaches take the ball first or second in OT? We say they should take it first -- but it's a close call and there are arguments for both sides.

From 1974 to 2011, overtime was sudden death, good and proper. First team to score wins. Winning the coin toss and taking the first possession was accordingly always optimal. Even after 2012, when the NFL adjusted the regular-season rules to give the second-possession team a drive after a first-possession field goal, it was still optimal to take the ball first, because a touchdown ended the game. Since 2012, there have been 201 overtime coin tosses -- only two winners elected to kick off, and only three winners deferred the choice to pick a side of the field to defend. There was rarely any thought to the decision.

Until now. In 2025, the NFL is adopting the postseason overtime rules they've used for the past few seasons during the regular season. A first-possession touchdown no longer ends the overtime period. Short of a defensive score (or a possession that lasts the entire 10-minute extra period), both offenses will now possess the ball. If, at the end of each first possession, the score remains tied, then the next score wins. Should the 10-minute overtime period expire with the score still deadlocked, the game ends in a tie.

We saw something close to this overtime format play out in Super Bowl LVIII between the Kansas City Chiefs and San Francisco 49ers. It wasn't exactly analogous, as playoff games can't end in a tie -- but the difficulty of the decision remains. There are advantages to taking the ball first and advantages to taking the ball second, and they aren't concrete. They change relative to both opponent team strength and opponent behavior. In the case of the Chiefs and 49ers, San Francisco's decision was heavily scrutinized after the team's overtime loss. But more significant than the specifics of the choice was the fact that few on the outside -- and even few among the 49ers -- totally understood all of the dynamics at play.

As such, ESPN surveyed analytics staffers from nine NFL teams with a series of questions about the new regular-season overtime rules for the 2025 season. This year marks the first regular season in which the overtime coin flip decision is a meaningful moment -- one in which sharp teams might find an edge. But as our survey shows, those edges still aren't totally clear.

Jump to:
Right coin-toss choice | Actual coin-toss choice
SBLVIII effect | First or second possession?
Other factors | Go for two? | Direction over ball?

When teams win the coin toss, what should they choose?

Responses:

• Lean first possession, but matchup factors could make second possession feasible (3 votes)

• Lean second possession, but matchup factors could make first possession feasible (3 votes)

• Almost always first possession (1 vote)

• Almost always second possession (1 vote)

• True 50-50 (1 vote)

Teams are split on this decision. The survey response above indicates that NFL analytics teams are coming to different conclusions about the best strategy. It's understandable, because there are countervailing forces at play, and each option offers a potential advantage.

The second possession team has the advantage of knowing what the first possession team did on its drive. If the first possession team scored a touchdown, the second possession team knows it must go for any and all fourth downs -- which can provide a playcalling advantage on third down, too, when the offense knows it doesn't have to reach the line to gain because it will have another shot. If the first possession team failed to score, the second possession team knows it can kick a short field goal, even on a fourth-and-1.

But the first possession team also has an edge: It receives the ball in any potential third possession and it can shorten the second possession with a long opening drive. If neither team scores in the first two drives, or the teams trade field goals, then the first-possession team gets the ball back in a true sudden-death scenario.

This is the kind of question where quantitative analysis helps. But as the results above show, different analysts can come up with different responses to the same question. So what does ESPN Analytics' model show?

The short answer is it favors taking the first possession, but there's more to it. In fact, a lot of the decision hinges on touchbacks. In another critical rule update, touchbacks are now coming out to the 35-yard line (up from the 30-yard line last season), and the expected rate of the touchback changes this analysis.

If touchbacks continue near the 64% rate they occurred last season, then our simulation model would estimate the first-possession team's net chances to win at 53.6%. But touchbacks will likely be less frequent given the yard-line change. If touchbacks drop to, say, a 40% rate, then we would estimate a 52.6% net win probability for the first-possession team. If it falls to 10%, then the first-possession team would have a 50.9% net win probability.

This is because having the ball at the 35 helps the first-possession team. If they are forced to punt or commit a turnover, it will be farther downfield -- thus lessening the probability of a second-possession score and improving field position for a third possession.

But what is truly relevant isn't the leaguewide average touchback rate; it's the touchback rates for a team and its opponent. And the team kicking off has significant control over whether there is a touchback. In other words, if a team chooses to kick, it can do so knowing it is trying to avoid a touchback (though this could cost it a couple of yards if the kick is short). Sharp teams that win the coin toss against touchback-heavy opponent kickers might elect to take the first possession, hoping to start way up at the 35.

Ultimately, if we were filling out the survey, we would respond with: "Lean first possession, but matchup factors could make second possession feasible." If a team knew it was going to avoid a touchback, it's close enough that additional context could push the optimal decision to second possession. But our analysis suggests that most of the time teams ought to opt for the first possession.

When teams win the coin toss, what will they choose?

Responses:

• Mostly (but not always) choose second possession (7 votes)

• Mostly (but not always) choose first possession (2 votes)

The disparity between these first two questions is interesting. Even some of the teams that believe that first possession is the correct choice also believe most teams will opt for the second possession.

Why could that be? We have one theory.

Do you think the result of the 49ers-Chiefs Super Bowl overtime will affect OT decision-making for teams in the 2025 regular season?

Responses:

• Yes (6 votes)

• No (2 votes)

• One team abstained

In Super Bowl LVIII, the 49ers won the overtime coin toss and chose first possession. You know the rest: The 49ers kicked a field goal on their possession, and the Chiefs responded by scoring a Super Bowl-winning touchdown.

Afterward, we learned that the Chiefs were shocked by the 49ers' decision to take the first possession. Even knowing the result, analytics staffers were split. But still, it's not hard to imagine there could be a lasting effect as head coaches try to avoid the heat coach Kyle Shanahan took.

Unlike in the postseason, regular-season overtime lasts only 10 minutes and can result in a tie. Compared to the postseason rule, does that favor first possession or second possession?

Responses:

• Second possession (5 votes)

• First possession (4 votes)

You can understand how teams could see this both ways.

The edge to the first-possession team is that it can slow-roll its opening drive. It's possible it can score on its first possession and leave the second-possession team only a couple of minutes to respond.

On the other hand, the shortened time frame means that the third possession -- the first-possession team's big advantage -- becomes less likely to happen because time could expire.

ESPN's model shows virtually no difference in net win probability between the playoffs and regular season. It moves ever so slightly toward second possession but close enough that we can't rule out it being just noise.

Which matchup factors could affect a first- vs. second-possession decision for teams?

Responses (Survey respondents were asked to check all that apply):

• Relative strength of offense vs. defense (8 votes)

• Weather (6 votes)

• Desire or lack thereof to go for two (5 votes)

• Desire or lack thereof to go for fourth downs (4 votes)

• Value of a tie for selecting team (4 votes)

• Fatigue on one side of the ball (2 votes)

One of the arguments against the 49ers' Super Bowl decision was that it gave Patrick Mahomes & Co. the informational advantage to go for fourth downs and, in a touchdown-touchdown scenario, the ability to go for two. That seems valid, and when teams feel the first and second possession choice is very close, considering each team's strengths could play a role.

The first-possession team will also have to tangle with potentially difficult and controversial fourth-down choices in an unusual game state. It will be weighing how much to risk to keep a non-sudden-death drive alive, with failure resulting in a sudden-death drive for its opponents. The second-possession team has much simpler fourth-down considerations since it knows what happened on the first drive. A team's confidence in its own ability to get those calls right could push its decision.

A surprise to us: Most teams did not select fatigue as being a driving factor. We figured a coach whose defense was gassed might want first possession to give his players a breather (or, conversely, take first possession when the opponent's defense is gassed). Perhaps this is just something the coach might factor in more heavily than analysts. Or maybe we overrated that dynamic.

What's the "go for two" strategy? This is straightforward. If the first possession team scores a touchdown, it ought to kick the PAT. If the second-possession team also scores a touchdown, it should go for two. Why? Because otherwise it would be giving its opponent a true sudden-death possession. There's virtually no argument to deviate from that plan. The only edge case is a situation in which a tie is atypically beneficial, but a loss would be detrimental -- as was the case for the Chargers in the infamous Chargers-Raiders Week 18 game in the 2021 season.

Will any team go for two after scoring a touchdown on the first possession?

Responses:

• No (8 votes)

• Yes (1 vote)

Why shouldn't the first team go for two? If it kicks the PAT and its opponent scores a touchdown, its win probability becomes simply the chance it can stop the opponent's 2-point attempt (roughly 50%, perhaps a shade higher). By comparison, if it goes for two and fails, it will lose almost all the time when its opponent scores a touchdown (because the opponent will know to kick a PAT). If the first-possession team goes for two and converts, it still won't win all of the time, because its opponent could also go for two and convert.

Will any team opt to play the wind and choose direction instead of first or second possession?

Responses:

• No (5 votes)

• Yes (3 votes)

• One team abstained

Because some teams view the first vs. second possession question as being extremely close, it's feasible they could view an external factor such as wind to be more important. In that case, they could choose which direction to play and defer the first vs. second possession question to their opponent.

This could be particularly meaningful early in the season -- when teams don't have a ton of data yet on how their opponents behave on kickoffs and can't accurately predict a touchback likelihood. But late in the season, when weather is poorer in open-air stadiums, it also becomes a factor.

ESPN's Brian Burke contributed analysis to this report.