How to bet on the NFL on Polymarket: moneylines, spreads, totals, props, and live games

Cowboys vs. Eagles opened the season on September 4, 2025 with 5 markets on Polymarket. Bills vs. Texans on September 13, 2026 listed 379, 35 of them spread lines, and a separate player-props event added 316 more. Count only the markets that traded 1,000 shares and an NFL game still went from 5.0 to 43.9 in a year, about 9 times as many.

The money did not spread out with the menu. On the 65 NFL games from September 10 to October 10, 2026, the moneyline took 54.1% of Polymarket's reported volume, the spread ladder 29.9% and the total ladder 14.4%. Team totals, halves, quarters and every prop shared the last 1.6%.

Polymarket titles each game with the away team first, "Bills vs. Texans" for the Bills at the Texans, and the moneyline pays $1 a share on the team that wins. Past the big three, a market can show a price with almost nothing behind it, so check its volume before you size a bet.

Packers vs. Cowboys finished 40-40 on September 28, 2025, and every share of its moneyline, 4.1 million shares of reported volume, paid 50 cents. It was the only tie among 348 regular-season and playoff moneylines that settled from September 2025 to October 5, 2026. The preseason, which has no overtime, produced 2 in 49: Colts vs. Patriots 13-13 and Seahawks vs. Chiefs 9-9.

Halves end level far more often than games. 14 of 134 first-half moneylines settled 50-50, and 7 of 22 quarter and second-half moneylines did. If you buy a half or quarter moneyline, a 50-cent payout is one of its ordinary outcomes, not a fluke.

A spread such as "Spread: Bills (-1.5)" resolves to the Bills if they win by 2 or more points and to the Texans otherwise, a tie included. A total such as "Bills vs. Texans: O/U 41.5" resolves Over at 42 or more combined points. A half-point line cannot land on the number, and none of 4,913 settled NFL spreads and totals split.

A postponed game keeps its markets open until it is played, and a game canceled with no makeup settles the moneyline 50-50. Props name their own sources and edge cases in the rules on each market's page.

A Polymarket share pays $1, so its price is a probability. A team at 65¢ is a 65% chance to win; the other team is about 35¢. Buy $100 of the 65¢ side and you hold about 154 shares that pay about $154 if the team wins.

Sportsbook odds convert the same way, and the chart above draws the conversion for the whole range. A favorite at -186 is 186 divided by 286, 65%; an underdog at +186 is 100 divided by 286, 35%.

A book's two sides add up to more than 100%. The two sides of a Polymarket market add up to about $1, and the cost of trading is the spread between the best bid and ask.

The kickoff price has been close to right. In 562 NFL games from September 2024 to September 2026 that traded $10,000 or more before kickoff, underdogs averaged 33.0¢ at kickoff and won 32.2% of the time, 0.80 points short of their price with an interval from minus 4.59 to plus 2.95 that includes zero. The football underdogs study has the favorites at each price and the biggest upsets.

NFL home teams won 53.7% of 566 regular-season and playoff games settled on Polymarket from September 2024 to September 2026, neutral-site games left out, with a half-width of 4.11 points, so the interval still includes an even split.

By calendar year it was 52.2% over 251 games in 2024 and 54.0% over 276 in 2025. The NFL's 53.7% ties MLB for the smallest home edge of the 6 US leagues in the home field study, where NBA home teams won 55.6% and college football home teams 64.4%.

Home advantage is in the price as well. Across 104,445 large Polymarket bets on home and away sides in every league measured, both sides came in within a point of what they paid. The NFL's own pricing sample is too small to read yet, 73 large buys, so treat home field as a small edge the market already counts, not a reason to back the home team.

Each rung of a spread ladder is its own market, so the price gap between two neighboring lines is the price of the margin between them. At 12:18 UTC on October 11, 2026, before Browns vs. Jets kicked off, Jets -2.5 traded at 51.5¢ and Jets -3.5 at 42.5¢: moving the Jets past 3 points cost 9 cents, against 3 to move them past 4.

The books were pricing football's scoring. 86 of 571 NFL games from September 2024 to September 10, 2026, 15.1%, ended on exactly 3 points, a field goal. Seven, a touchdown and its extra point, came next at 8.1%, and the Jets ladder charged 5 cents for it, no more than for a 6. One ladder on one morning is too thin to price 7; 571 games are enough to pin 3.

The point spread guide covers spreads across sports. On an NFL ladder, the half-point either side of 3 is the one worth paying for: -2.5 rather than -3.5 when you lay the points, +3.5 rather than +2.5 when you take them.

Half and quarter markets look like a way to bet a smaller slice of the game. On Polymarket they mostly mean a thinner book: from September 10 to October 10, 2026, 15 first-quarter spreads traded 1,000 shares or more, against 1,093 full-game totals. An order the game total would absorb can move a quarter line's price against you before it fills.

They are noisier bets too. A quarter is a quarter of the playing time, so one drive decides more of the result, and a level half or quarter splits its moneyline 50-50. Team totals work like the game total for one team's points.

The game total is a ladder like the spread, one market per line, each paying Over or Under on the combined score, and the over/under guide covers how totals are priced. For an NFL total, trade the full-game ladder and choose your line there.

NFL teams name their inactive players 90 minutes before kickoff, and the large money comes late. Of 2,771 NFL buys of $10,000 or more from September 10 to October 10, 2026, 62.1% landed in the last hour before kickoff or during the game, and only 16.9% six hours or more ahead.

So a price a day out has met little of the money it will meet, and an in-play price moves with every possession. To take a position off mid-game you sell your shares, and the cash-out guide covers when that makes sense. Either way, a limit order at your price keeps you from chasing a moving line, and a resting order that fills pays no fee: Polymarket charges only the taker.

From September 20 to October 10, 2026, the busiest NFL game drew 145 large buys in 0xinsider's large-trade feed. The graded money on NFL games split almost evenly: wallets graded S, A or B when their order filled placed 44.6% of 2,461 large NFL buys, and D and F wallets 44.3%.

The NFL board lists the next 7 days of NFL games with an active Polymarket market, live games first. Each card shows both teams' prices and chances, the 24-hour move, and the quarter and score for a live game. Under Details, free on every card, sit the most traded props and the top 5 holders on each side with their grades and football win rates.

On the event page, volume by market family and the full Top holders list are Pro, shown as labeled sample data without it. The large-trade feed lists recorded taker fills of $10,000 or more that meet its market-volume-share floor, marked BUY or SELL with the wallet's grade, live with Pro and 24 hours later without it. The sharp money guide walks through reading one game's wallets.

A grade tells you who moved a price, not who will be right. Across 51,797 large sports buys 0xinsider recorded from June 1 to September 11, 2026, S, A and B wallets came in +0.50 points against the price they paid, an interval from -0.58 to +1.59 that includes zero, and D and F wallets -0.18, from -1.73 to +1.36. Read a large trade as news about the book, and price the game yourself.