The NFL Schedule Formula: How Each Team's Games Are Chosen
Every NFL team's opponents follow a fixed formula tied to divisions and last year's finish. Here is how the 17 games are built, why schedules differ, and what is left for the league to decide.
Every spring the NFL releases its schedule, and every spring fans notice the same thing: their team's opponents were known months earlier, and some teams seem to get an easier road than others. Who plays whom is not a mystery and is not picked by a committee. It is the output of a formula that has been in place, with adjustments, for many years.
The league does make real choices, but they are about when and where games are played, not about who plays. Dates, kickoff times and television slots are fitted around the matchups that the formula has already decided. If you understand the formula, you can work out most of a team's opponents for the coming year on your own, before the official release.
This article builds a 17-game schedule one piece at a time, using a hypothetical team for the examples. The formula is written for the current 17-game season. It has been revised when the league changed the length of the season and could be revised again, so confirm details with the league's published schedule notes.
The building blocks
The league has 32 teams. They are split into two conferences of 16, and each conference has four divisions of four teams. That structure is the skeleton of the whole formula. Each team's 17 games break down into five groups:
| Group | Games | Opponents | Home/away |
|---|---|---|---|
| Own division | 6 | Each of the three division rivals, twice | 3 home, 3 away |
| Same conference, rotating division | 4 | All four teams in one division of your own conference | 2 home, 2 away |
| Other conference, rotating division | 4 | All four teams in one division of the other conference | 2 home, 2 away |
| Same conference, same-place finish | 2 | The team that finished where you did last year in each of the two remaining divisions of your conference | 1 home, 1 away |
| Extra game | 1 | A same-place finisher from a division in the other conference | Home or away, alternating |
Add the games together and you get 6 + 4 + 4 + 2 + 1 = 17. Every piece is predictable except for one input: where each team finished the previous year. Let us take them one at a time.
Six games inside the division
Each team plays every other team in its division twice, once in each stadium. These six games never change from year to year, and they are the reason a division title is such a consistent measure of a team: everyone in the division has the same six opponents. They are also the games that give the league its best rivalries, so they are the ones most likely to be placed in prime time and, in recent seasons, the final week of the regular season.
Because every team's six division games are home-and-home, home and road are perfectly balanced there. Any imbalance in the full 17 can only come from the other groups.
Four games against a division in your own conference
Here the formula starts rotating. Each conference has four divisions, so each team has three other divisions in its conference. The rotation matches your division against one of those three each year, and the pairings step through all three in order, which takes three years to repeat. In the year your division is paired with another, every team plays every team in the other division: four opponents, two at home and two away.
A concrete example, with a hypothetical rotation: say this year the East division of a conference is paired with its South division. Then every East team plays all four South teams, and every South team plays all four East teams. The East's other two matching groups, the North and the West, are on the schedule for the next two years. After three years every team in a conference will have played every other team in the same conference at least in the fixed rotation.
The pairing is set in advance for years ahead and does not depend on how well anyone played. If you want to find out who your team hosts in the rotation game this year, you only need to know which division it is paired with.
Four games against a division in the other conference
The same idea applies across conferences, but because the other conference has four divisions and your division can be paired with one of them each year, the cycle is four years long. In a given year a division faces one entire division from the other conference, again two home games and two on the road.
The result is that a team plays each opponent from the other conference at least once over a four-year cycle in the rotation. Combined with the three-year cycle for the other games, the pairings of whole divisions line up every 12 years. That is a long enough time that fans of one team can go a decade without seeing another franchise visit their stadium, which is part of why a visit from a team in the other conference can feel like an event.
Two games based on last year's finish
This is the part of the formula that depends on performance. After the rotating division is chosen, each conference still has two divisions that the team is not paired with in the four-game set. For each of those, the team plays the club that finished in the same position in the standings the previous year. A first-place team plays the first-place teams of the two remaining divisions, and a last-place team plays the two last-place clubs.
The point of this part of the formula is balance. It matches strong teams with strong teams and weak teams with weak ones, so that a champion has to continue to beat good opponents while a struggling team gets a slightly softer path back to respectability. Over a full season the effect is limited to two games out of 17, but it is the reason published "strength of schedule" figures differ from team to team.
One of the two games is at home and the other is on the road. As with the other pieces, the formula decides which, so a team that finished first and one that finished last do not have different home and away counts.
The 17th game
When the league moved to a 17-game schedule it needed one more game for each team, and it needed a way to add a game without breaking the structure. The answer was an extra matchup against a team from the other conference. The opponent is the team that finished in the same position as you in a division of the other conference. It is not a division your division is already paired with that year, which would mean playing the same team twice.
Because 32 teams cannot all host an extra game, the host alternates: in one year one conference is designated to host the extra game, and the next year the other conference is. As a consequence, some teams play nine home games and eight road games in a given year, and others play eight home games and nine on the road, with the imbalance switching sides the following season. This also explains the odd feeling of a team that has more home games than usual getting a longer string of home dates.
Reading a team's schedule backwards
You can use the table above to reconstruct a team's whole year. Take a hypothetical team that finished second in its division last year:
- Division games: the three division rivals, home and away. That is six games.
- Same-conference rotation: the four teams of the paired division, two home and two away.
- Other-conference rotation: the four teams of the division the league paired with yours this year.
- Same-place games: the second-place team in each of the two remaining divisions of your conference.
- Extra game: the second-place team in a division of the other conference that is not already one of your opponents.
If you add them up you get 17 opponents, and with the division games counting twice, 14 different teams. That is a quick way to check any published schedule for mistakes.
What the league actually decides
Once the matchups are fixed, the league has to turn a list of 272 games (at 17 per team, 32 teams, each game counted once) into a calendar. That is a scheduling puzzle with a lot of constraints, and it is solved with specialized computer software and then reviewed and adjusted by league staff. The inputs include:
- Stadium availability: many NFL teams share a venue with other events, and some stadiums are not available on particular dates. Baseball and other events, concerts and weather considerations have to be handled.
- Broadcast contracts: the networks and streaming partners that carry games have agreed to a set of windows, and the league has to provide enough high-profile games for each one. See How NFL Games Reach Your Screen.
- Bye weeks: every team gets one, placed so that no team has it too early or too late and no division is lopsided in how many teams are off in one week.
- Rest and travel: the league tries to avoid long runs of consecutive road games and unusually short weeks, although with Thursday games on the calendar it cannot eliminate short weeks entirely.
- International games: the league plays some regular-season games abroad, and the home team for those games gives up a home date. See International Games: How the NFL Plays Abroad.
- Late-season rules: in recent seasons the final week of the regular season has been reserved for division games, so that teams chasing the same title are playing games that matter to each other.
The league also retains flexibility on the back end. Some Sunday games are moved to the Sunday night slot late in the season based on what is at stake, a practice explained in Flexible Scheduling: How Primetime Games Are Chosen. So the date you see in May for a late-season game may change, but the opponents will not.
Why schedules still feel unfair
Three factors make a formula-based schedule feel uneven even though it is symmetric on paper.
The same-place games depend on last year. A team that finished first is matched with other first-place teams, but a team that finished first last year may not be a first-rate team this year, and a last-place team may have improved. Published strength-of-schedule figures usually use last year's records for exactly that reason, which means they say something about the previous season more than about the coming one.
Divisions differ in strength. A team in a strong division plays six games against strong opponents. A team in a weak division plays six games against weaker ones. The formula does not correct for this because the division games are fixed.
Rest is not a formula item. The calendar can give a team a short week, a long week, a trip across three time zones or a late bye. These are real effects, but they are decided in the scheduling stage and not by the matchup formula.
What it means for the standings
Because the six division games are the same for everyone in the division, the first tiebreaker steps (head-to-head, then division record) are fair across the teams that are being compared. Outside the division the opponents differ, which is why the later tiebreakers look at common games and strength of schedule. See How NFL Playoff Tiebreakers Actually Work for the order. You can follow the schedule page for upcoming games and the standings page for the live table, and each team's page shows its own list of opponents.
The short version
- Opponents are set by formula; the league decides dates, times and television slots, not matchups.
- Each team plays 17 games: 6 in its division, 4 against a rotating division in its own conference, 4 against a rotating division in the other conference, 2 against same-place finishers in the other two conference divisions, and 1 extra interconference game.
- The same-conference rotation repeats every three years and the other-conference rotation every four.
- Only the same-place games, and the extra game, depend on last year's finish.
- Nine home games and eight road games alternate between conferences each year because of the 17th game.
- Check the league's schedule notes for the current year's pairings, since the formula was written for the 17-game season and could be revised.