Simulating the 2025 NFL Season
After 25,000 simulations of the season, which teams are projected at the top of the standings?
I've already posted my ratings heading into the NFL season, with the Baltimore Ravens as the clear favorite and the Carolina Panthers in last by a significant margin. But I have one more article to write before kickoff this evening, and that's a preview of where the 32 teams will end up at the end of the season. In the absence of actual regular season action like in the photo, I decided to run some computer-generated simulations of the NFL season to predict the final standings.
I wrote a program in Python to randomly simulate the NFL season using my team ratings, and I ran 25,000 simulations. Compared to the time required to calculate the team ratings, this actually runs surprisingly fast across the four cores of my old quad-core Xeon. The simulator predicts the score of each game using the offense and defense ratings, then it adds in a random adjustment to account for the uncertainty in predidcting the games. When I introduced my rating system, I mentioned that results I post are actually the median of many slightly different rating attempts. For NFL ratings, I actually run 100 different attempts, and the results you see are the median of those attempts. For this simulation, I varied team team's offense and defense ratings in each simulated season using the results from the individual rating attempts. This doesn't seem to make too much of a difference, but perhaps it adds a bit more variation in the simulations.
After each season is simulated, I apply simplified tiebreakers as needed to determine the top seed in each conference, division winners, and wild cards. The real NFL tiebreaking rules are somewhat complicated, so I've used a simpler version that works better with my simulator. If there are two teams tied for a position, the first tiebreaking attempts are based on head-to-head record and point differential. If the teams are still tied after that, or if three or more teams are tied, I apply a series of tiebreakers to eliminate the poorest performing team. And I keep eliminating teams from the tie until I’m left with a single team. For a division title, the tiebreakers are the winning percentage and point differential in divisional games, the winning percentage and point differential in conference games, overall point differential, and finally random chance. For the top seed in each conference and the wild cards, I skip the winning percentage and point differential in divisional games, and I skip to the remaining tiebreakers. It's a simplified approach, but implementing a more detailed and accurate tiebreaker probably won't affect the overall results very much after 25,000 simulations.
My simulator doesn’t include postseason games yet, but I might add that in the future. It wouldn’t be that hard to add this functionality, but I’m not sure I have time to do that before tonight’s game. Or perhaps I’ll just blame Jim Mora for instructing fans to not talk about the playoffs. I’ll be releasing the code for my ratings in a few days on Github, and I expect to release my simulator about a week later. I’m aiming to add postseason simulations by the time I release the code to the simulator, but I make no guarantees.
Anyway, let's take a look at the results.
Simulated Standings
The PF and PA columns are the mean points for and points against. In all three tables, teams are sorted according to their probability of making the playoffs. The rightmost column has each team's overall rating.
AFC East
W L T Win% PF PA Rating
Buffalo Bills 11.09 5.85 0.06 .654 478.40 379.07 5.28
Miami Dolphins 8.77 8.16 0.07 .518 356.68 346.18 0.27
New York Jets 7.95 8.98 0.07 .470 355.50 375.20 -1.85
New England Patriots 6.51 10.42 0.06 .385 332.04 405.29 -5.23
AFC North
W L T Win% PF PA Rating
Baltimore Ravens 13.32 3.63 0.05 .785 487.10 300.32 11.86
Pittsburgh Steelers 8.51 8.42 0.07 .503 349.74 349.13 2.04
Cincinnati Bengals 8.20 8.74 0.07 .484 433.98 444.86 1.15
Cleveland Browns 5.71 11.21 0.07 .338 285.73 387.80 -4.62
AFC South
W L T Win% PF PA Rating
Houston Texans 9.39 7.54 0.07 .554 351.85 319.13 2.43
Indianapolis Colts 9.08 7.85 0.07 .536 389.89 367.61 0.47
Jacksonville Jaguars 6.14 10.80 0.06 .363 326.85 410.91 -5.82
Tennessee Titans 5.76 11.17 0.07 .341 304.85 402.75 -6.67
AFC West
W L T Win% PF PA Rating
Denver Broncos 11.88 5.06 0.06 .701 419.69 293.01 7.34
Los Angeles Chargers 10.85 6.09 0.07 .640 393.38 306.54 5.24
Kansas City Chiefs 8.56 8.38 0.07 .505 382.60 379.33 1.62
Las Vegas Raiders 5.73 11.21 0.07 .339 300.49 401.90 -5.62
NFC East
W L T Win% PF PA Rating
Philadelphia Eagles 11.01 5.93 0.06 .650 435.66 342.10 6.51
Washington Commanders 7.71 9.23 0.07 .455 412.80 440.31 -0.75
New York Giants 7.54 9.39 0.07 .445 343.78 377.49 -1.53
Dallas Cowboys 5.55 11.39 0.06 .328 341.64 449.68 -5.54
NFC North
W L T Win% PF PA Rating
Green Bay Packers 10.11 6.82 0.07 .597 406.15 344.31 4.83
Detroit Lions 9.69 7.24 0.07 .572 437.39 393.78 4.47
Minnesota Vikings 8.67 8.26 0.07 .512 364.08 357.25 2.12
Chicago Bears 8.19 8.74 0.07 .484 347.70 358.08 0.52
NFC South
W L T Win% PF PA Rating
Tampa Bay Buccaneers 11.72 5.22 0.06 .691 450.67 325.98 5.25
New Orleans Saints 6.90 10.03 0.07 .408 326.01 382.75 -5.44
Atlanta Falcons 6.58 10.35 0.07 .389 343.51 411.77 -5.58
Carolina Panthers 3.62 13.32 0.05 .215 298.29 490.46 -12.63
NFC West
W L T Win% PF PA Rating
Seattle Seahawks 9.54 7.39 0.07 .563 384.58 345.11 0.79
Arizona Cardinals 9.45 7.48 0.07 .558 380.29 344.59 0.48
Los Angeles Rams 8.82 8.12 0.07 .521 386.66 373.73 0.36
San Francisco 49ers 8.39 8.54 0.07 .496 392.90 394.49 -1.86 Playoff Probabilities
I've also calculated each team's probability of making the playoffs, winning their division, getting the top seed in their conference, and the team’s mean seed for all of the simulations in which the team actually did reach the postseason.
AFC East
Rating Win% Playoff% Div% Conf% MeanSeed
Buffalo Bills 5.28 .654 86.60% 70.86% 9.92% 3.31
Miami Dolphins 0.27 .518 44.32% 17.64% 0.94% 4.94
New York Jets -1.85 .470 27.26% 8.92% 0.37% 5.27
New England Patriots -5.23 .385 9.24% 2.58% 0.02% 5.60
AFC North
Rating Win% Playoff% Div% Conf% MeanSeed
Baltimore Ravens 11.86 .785 98.95% 93.23% 55.73% 1.83
Pittsburgh Steelers 2.04 .503 39.90% 3.84% 0.87% 5.68
Cincinnati Bengals 1.15 .484 33.37% 2.72% 0.59% 5.79
Cleveland Browns -4.62 .338 4.48% 0.21% 0.01% 6.24
AFC South
Rating Win% Playoff% Div% Conf% MeanSeed
Houston Texans 2.43 .554 64.36% 50.60% 2.33% 3.96
Indianapolis Colts 0.47 .536 56.89% 41.63% 1.42% 4.18
Jacksonville Jaguars -5.82 .363 8.22% 4.54% 0.04% 4.98
Tennessee Titans -6.67 .341 6.10% 3.23% 0.00% 5.09
AFC West
Rating Win% Playoff% Div% Conf% MeanSeed
Denver Broncos 7.34 .701 93.26% 60.42% 18.89% 3.27
Los Angeles Chargers 5.24 .640 82.90% 32.76% 7.79% 4.28
Kansas City Chiefs 1.62 .505 39.85% 6.50% 1.06% 5.50
Las Vegas Raiders -5.62 .339 4.30% 0.32% 0.01% 6.18
NFC East
Rating Win% Playoff% Div% Conf% MeanSeed
Philadelphia Eagles 6.51 .650 89.45% 79.96% 22.62% 2.81
Washington Commanders -0.75 .455 27.89% 9.99% 0.93% 5.01
New York Giants -1.53 .445 24.98% 8.73% 0.76% 5.07
Dallas Cowboys -5.54 .328 4.56% 1.33% 0.04% 5.48
NFC North
Rating Win% Playoff% Div% Conf% MeanSeed
Green Bay Packers 4.83 .597 74.14% 40.79% 10.52% 3.93
Detroit Lions 4.47 .572 65.74% 32.35% 8.23% 4.17
Minnesota Vikings 2.12 .512 45.46% 16.02% 2.86% 4.80
Chicago Bears 0.52 .484 35.26% 10.83% 1.60% 5.03
NFC South
Rating Win% Playoff% Div% Conf% MeanSeed
Tampa Bay Buccaneers 5.25 .691 96.27% 93.27% 34.57% 2.32
New Orleans Saints -5.44 .408 15.28% 3.77% 0.29% 5.42
Atlanta Falcons -5.58 .389 11.77% 2.90% 0.21% 5.47
Carolina Panthers -12.63 .215 0.25% 0.06% 0.00% 6.00
NFC West
Rating Win% Playoff% Div% Conf% MeanSeed
Seattle Seahawks 0.79 .563 61.95% 33.10% 6.27% 4.19
Arizona Cardinals 0.48 .558 60.34% 31.16% 5.90% 4.24
Los Angeles Rams 0.36 .521 48.39% 20.83% 3.20% 4.62
San Francisco 49ers -1.86 .496 38.27% 14.91% 2.03% 4.83Potential Outcomes
To show the range of possibilities for each team during the season, I calculated the probability of a winning season (>.500 winning percentage) and several percentiles for each team's winning percentage. These are intended to show more of the typical, best case, and worse case scenarios for each team.
AFC East
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Buffalo Bills .654 91.51% .529 .588 .647 .706 .794
Miami Dolphins .518 55.90% .353 .441 .529 .588 .647
New York Jets .470 38.33% .324 .412 .471 .529 .618
New England Patriots .385 15.06% .235 .294 .382 .471 .529
AFC North
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Baltimore Ravens .785 99.58% .647 .706 .765 .882 .882
Pittsburgh Steelers .503 50.02% .353 .412 .529 .588 .647
Cincinnati Bengals .484 43.86% .353 .412 .471 .588 .647
Cleveland Browns .338 7.30% .176 .235 .353 .412 .471
AFC South
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Houston Texans .554 67.50% .412 .471 .529 .647 .706
Indianapolis Colts .536 61.65% .412 .471 .529 .618 .706
Jacksonville Jaguars .363 10.92% .235 .294 .353 .412 .529
Tennessee Titans .341 7.81% .176 .235 .353 .412 .471
AFC West
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Denver Broncos .701 96.17% .559 .647 .706 .765 .824
Los Angeles Chargers .640 89.04% .471 .588 .647 .706 .765
Kansas City Chiefs .505 51.08% .353 .412 .529 .588 .647
Las Vegas Raiders .339 7.34% .176 .235 .353 .412 .471
NFC East
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Philadelphia Eagles .650 90.30% .529 .588 .647 .706 .765
Washington Commanders .455 34.52% .294 .353 .471 .529 .588
New York Giants .445 31.32% .294 .353 .471 .529 .588
Dallas Cowboys .328 6.18% .176 .235 .324 .412 .471
NFC North
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Green Bay Packers .597 79.84% .471 .529 .588 .647 .765
Detroit Lions .572 72.41% .412 .471 .588 .647 .706
Minnesota Vikings .512 53.31% .353 .412 .529 .588 .647
Chicago Bears .484 43.57% .353 .412 .471 .588 .647
NFC South
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Tampa Bay Buccaneers .691 95.63% .529 .618 .706 .765 .824
New Orleans Saints .408 20.68% .235 .353 .412 .471 .529
Atlanta Falcons .389 16.11% .235 .294 .412 .471 .529
Carolina Panthers .215 0.30% .118 .147 .235 .294 .353
NFC West
Win% >.500% 10%ile 25%ile 50%ile 75%ile 90%ile
Seattle Seahawks .563 70.03% .412 .471 .588 .647 .706
Arizona Cardinals .558 68.70% .412 .471 .559 .647 .706
Los Angeles Rams .521 56.64% .353 .471 .529 .588 .647
San Francisco 49ers .496 47.74% .353 .412 .471 .588 .647Conclusions
I'll admit that this is a somewhat brief article just because I spent so much time coding and testing the simulator, and I want to get this posted while it’s still early in the day. But there are some interesting results to discuss before posting this article.
Does Baltimore really have a 98.95% chance of making the playoffs? That would be almost a sure thing, and it would be a very bold prediction to make before a single regular season game has been played. The answer to the question is almost certainly no, the Ravens’ probability of making the playoffs probably isn’t quite that high. My ratings and the 25,000 simulations take into account a team's expected production now, based on last season and preseason games this year. But that doesn't account for the potential for injuries, other roster moves that might occur during the season, players returning from being injured last year, the development of young players, and some other factors as well. There's definitely more uncertainty in the season than what my simulator accounts for. Just like I don’t completely trust the bold prediction about Baltimore, teams at the bottom of my ratings like Carolina might well exceed these projections. It's very possible the Panthers will outperform expectations due to the development of other players, or if other teams in their division are weaker than expected.
The NFC North is a really intriguing division because all of its teams have positive ratings. In a weaker division, even the Bears might have a greater than 50% chance of making the playoffs. The top of the NFC North isn't projected to be as strong as in some of the other divisions, but that could change if the Lions or Packers are better than expected. Detroit was ranked at the top of the final 2024 ratings, Green Bay was #5, and Minnesota was #7. If those teams are stronger than the preseason ratings suggest, the NFC North would look a lot like the SEC of the NFL in that would be easily the strongest division from top to bottom.
The NFC East is also interesting because the Eagles are projected as the clear favorite, but that depends quite a bit on the strength of the Washington Commanders. Washington was a surprise last season and they’ve dropped considerably from the final ratings last season. If the Commanders are better than expected, that could be a more competitive division. The Giants have also surged in the ratings this preseason and are an intriguing team. Neither team has a particularly high chance of making the playoffs according to my simulations, but I’m not totally convinced that Philadelphia will run away with this division.
The NFC West is easily the most balanced division, but they don't have the overall strength of the NFC North. Since I compared the NFC North to the SEC, the NFC West is a bit like the Big 12 in that there aren't any elite teams, but there also really aren't any truly bad teams. To a much greater extent than any other division, the NFC West is wide open for any team that outperforms their rating a bit, and that also makes it really interesting. The Seahawks average just 1.15 more wins than the 49ers, who are predicted to be at the bottom of the division.
I also want to talk about the Chiefs, who I also discussed when I posted my initial 2025 ratings. It might seem shocking that my simulator gives them less than a 50% chance of making the playoffs, but I don't think that's unreasonable. The AFC West has two other strong teams in the Broncos and Chargers, so it's going to be difficult for Kansas City to win the division. The simulated season is based on my computer ratings, which are derived from the point differential in games. Despite a 15-2 record last season, the Chiefs weren't dominating opponents. The same can be said for 2023, where their point differential was also underwhelming, and they had an 11-6 record. The Chiefs haven't been nearly as explosive on offense the past two seasons as they were in 2022, and that's cut into their rating quite a bit. They had a good record, but it’s hard to continue winning a lot of close games. If they played in the NFC West, the Chiefs would have a pretty good chance of making the playoffs, but it's a much more difficult path in the AFC West. They also play one of the toughest schedules overall in the NFL. The Chiefs having less than a 50% chance of reaching the postseason might seem like a surprising prediction, but it actually makes sense to me. I have no idea who will actually win the NFC West, but I’m much more confident in this prediction about the Chiefs.
The first of 272 regular season games takes place this evening, and there will certainly be surprises that defy my preseason ratings and simulations of the upcoming season. But I hope that my ratings at least have some skill at identifying the NFL's best teams and predicting the winners of those 272 games plus the postseason. These computer ratings began as an experiment a couple of years ago to predict the winners of college football bowl games, and I’m still continuing to tweak the algorithm and make improvements. No matter what happens over the next 18 weeks, it’ll be a fun ride, and I’m sure to learn a lot about what my rating system does well and where it can be improved.
Thanks for reading!