Line movement explained: how a price moves from open to kickoff to the final whistle

On October 11, 2026, Polymarket priced the Jets to cover -2.5 against the Browns at 51.5¢ and to cover -3.5 at 42.5¢. The 9¢ between them is the market's estimate that the Jets win by exactly 3, about 1 game in 11. No other half-point on the ladder cost more than 5¢.

At a sportsbook a bet has two numbers. The line is the spread or total, Chiefs -3.5 or over 47.5, and the price is what either side costs, usually -110. When money moves the market, the book usually moves the line and holds the price: -3.5 becomes -4 at the same -110. Sometimes it holds the line and moves the price, -3.5 at -115. A moneyline has no spread, so only its price moves.

On Polymarket the line is fixed inside each market. Jets -2.5 is one market and Jets -3.5 is another, and the chart above is 11 of the Jets' spread markets read in the same 25 seconds. What moves is the price in cents, and the price is the probability.

So a half-point is worth exactly what the market thinks it is, and you can see it. At a sportsbook you pay a set premium to buy a half-point; here each one has its own price, and the one across 3 cost the most. Every cent the -3.5 contract moves is a point of probability, with every trade behind the move on a public ledger and a wallet attached.

More than a week before kickoff, the average quoted spread on a large Polymarket sports buy was 21.2¢, and the median was 2¢. Most early books were tight and a few were close to empty: an early price is one almost nobody has checked yet. Look at the book before you trust it.

Money moves a line. A sportsbook that takes too much on one side moves the line to attract the other, and it moves further and faster for bettors it respects. An order book moves for a simpler reason: a buyer takes the offers at 48¢ and 49¢, and the next offer is 50¢.

Information moves a line. A starting pitcher is scratched, a quarterback is listed doubtful, the lineups drop an hour before a soccer match, the weather turns. The first to act on the news get the old price, the line moves to the new one, and their edge is the distance it moved.

Time moves a line. The closer to kickoff, the more information is in and the more money has voted, so the price converges and the spread narrows; the chart above is that narrowing.

And the game moves the line. Once play starts, a goal, an injury or a run of outs moves the price more in a minute than the whole pre-game week did. A sportsbook reprices in-play with a delay and a wider hold. A Polymarket market keeps trading on the same book.

Steam is a sudden, sharp move in the same direction across many sportsbooks at once. It happens when a bettor or a syndicate the books respect hits the same side everywhere within minutes, and every book moves to protect itself. Chasing steam means betting the side that moved, on the theory that whoever moved it knows something. The problem is that by the time you see it, you are paying the new price.

Reverse line movement is a line that moves against the side most of the tickets are on. If 70% of bets are on the home team and the line moves toward the away team, the money on the away side is bigger and more respected than the tickets on the home side. Reverse line movement is the classic sign that the public and the sharp bettors disagree, and that the book is siding with the sharps.

Both are ways of inferring who is behind a move from the shape of the move, because at a sportsbook the bettors are invisible. On Polymarket they are not. Every large buy is a wallet with a settled record, and 0xinsider grades that record. The equivalent of reading reverse line movement is looking at which side the profitable wallets are on, which the Profitable wallets block on each sports board shows directly.

The chart above is a real move that looked like steam. A week before the Seahawks played the Commanders on September 27, 2026, the Seahawks' price jumped from 57¢ to 72¢ in two hours, while both teams were playing their previous games: the Seahawks beat the Cardinals and the Commanders lost to the Cowboys. The price kept rising to 77.5¢ by kickoff, and the Seahawks lost. Anyone who chased the move paid up to 77.5¢ for a share that settled at zero. One game proves nothing either way: across the 60 NFL games of September 10 to October 10, 2026 whose price moved half a cent or more in the final week, the move pointed at the winner in 32 and away from it in 28.

0xinsider timed 401,482 large Polymarket sports buys (the smallest about $3,000 until July 2, 2026 and $10,000 after) against kickoff, across every sport that settled yes or no between April 2 and September 11, 2026. Half of them, 48.6%, landed after the game had started.

The last hour before kickoff took 22.8% and the first hour of play 22.4%, so 4 in 10 large sports bets land in the 2 hours around the opening whistle. Only 12.6% landed more than 6 hours out, and a tenth of a percent more than a week out.

The bets grow as kickoff approaches. A week out the average large buy was $10,799. In the last hour it was $22,375. The largest average stakes, $23,274, landed more than 2 hours into play, on long games and second-half markets. A sportsbook sees the same curve from the other side, thin when the line opens and flooding in the hour before the game.

The full numbers are in the timing study. For reading movement, the point is that most of the information a price will ever carry arrives in a narrow window around kickoff, and that a Polymarket price keeps absorbing large buys well into the game.

The received wisdom at a sportsbook is that the closing line is the sharpest and that early lines are soft, which is why professionals bet openers. The Polymarket data agrees about the sharpness and disagrees about who benefits.

Buys placed 1 to 7 days before kickoff won 57.4% of the time at an average price of 59.6¢: minus 2.19 points, with a 95% interval from minus 4.01 to minus 0.43. Buys placed more than a week out came in at minus 2.03 points on a smaller sample.

From a day before kickoff to the final whistle, every timing bucket was calibrated: within a point of zero, with an interval that includes it. The large buyer who waited for the day of the game paid a fair price. The large buyer who bet a week early paid a 21¢ spread on average to hit a price that had not yet absorbed the lineups, and in this sample was wrong more often than that price said.

So early lines are soft on Polymarket too, and the softness has gone to whoever was on the other side of the early money. If you have information a week out, the price is there to be taken. If you do not, the data says wait.

To bet at all you cross the spread, and the spread is part of the price. On Polymarket sports markets the quoted spread at the moment of a large buy was 1¢ at the median in every sport, and in every timing bucket inside the final week before kickoff, from April through July 2026. The 90th percentile was 1¢ in soccer, the NBA, baseball and hockey and 2¢ in tennis, esports, MMA and the WNBA.

A taker at 50¢ crossing a 1¢ book pays half a cent above the midpoint, 1% of the stake, plus Polymarket's sports taker fee of 2.5% of the stake at that price. A resting limit order pays no fee and collects the spread instead.

Against a sportsbook's 4.55% hold at -110, a Polymarket taker pays about 3.5% and a maker nothing, as long as the book is tight. The order book guide shows how to check the spread before you trade. A week before kickoff, check it twice.

Before kickoff most large buys are near a coin flip: in the last hour before the game 55.1% of buys were between 40¢ and 60¢ and 9.4% were at 80¢ or more. In the first hour of play the 40-to-60 share falls to 35.9% and the 80-plus share rises to 21.6%. In the second hour a third of all large buys are at 80¢ or more. A team is ahead, and the money is on it.

Those live prices were fair. In-play buys at 70¢ and up paid 84.1¢ and won 84.3% of the time. In-play buys between 30¢ and 70¢ came in at plus 0.67 points. The live longshot was the one exception: buys under 30¢ during play won at their price but returned minus 8.0% on the dollar, because the stakes that lost were larger than the stakes that won. Buying the team that is behind is cheap for a reason.

Which sports trade live is a fact about the sport. Cricket buys landed in-play 90% of the time, tennis 74%, MMA 74%, esports 66%. Soccer was 36%, hockey 31%, the WNBA 27%. In every sport with both phases, the in-play price was fairer to the buyer than the pre-kickoff price. Tennis buys before the first serve came in at minus 0.83 points; tennis buys during the match at plus 0.57. The tennis guide covers the tennis markets and live tennis.

From June 2026, when 0xinsider's grade coverage widened, every large buy also carries the grade its wallet held when it traded. Wallets graded S, A or B placed 51.4% of their sports buys in-play. Wallets graded D or F placed 37.7% in-play and 58.3% on the day before kickoff. The well-graded wallets trade about half live; the poorly graded wallets mostly trade before the game.

Before kickoff the grade does not separate anyone: in the last hour before the game, S, A and B wallets came in at -1.08 points and D and F wallets at +0.32. In the first hour of play, S, A and B wallets came in at +2.23 points and D and F wallets at -1.70, 3.93 points apart on the same games in the same hour.

The in-play gap is the one finding of the sports grade study that held up when it was re-measured with grades taken at the trade.

For a trader reading movement, the grades answer the question steam and reverse line movement try to guess at. On Polymarket the price moves during the game because a goal or an injury changed it, and the wallets that were right about the change are, in this data, the wallets that had been right before. When the Profitable wallets bars on a game move during play (live with Pro, on a delay without Pro), those wallets moved them.

Among games with at least one recorded large buy, the median NFL game drew 36 large buys in 0xinsider's feed from September 20 to October 10, 2026. The median soccer game in that selected population drew 2, and the median tennis match with a recorded buy 1 or 2. On most games the feed shows you a handful of trades, not a moving market, so read the count before you read the move.

Each game on a 0xinsider sports board, the NFL, NBA, soccer, tennis and the rest, shows the moneyline, spread and total prices, the kickoff countdown, and the live score once play starts. The Profitable wallets block on each game shows which side the profitable wallets are on and how much they have on it, live through the game with Pro and on a delay without.

The large-trade feed is the recorded large taker buys that pass its size and market-share filters, timestamped, with the buyer's grade beside it, live with Pro and 24 hours later without. In the hour before kickoff it shows the money the timing study calls late; in the first hour of play, the money that separates the grades. Pick of the Day records each pick's entry price and, where captured, its closing price, which is closing line value on a public record.

A price that moved on 36 large buys has been argued over. A price that moved on one has been pushed.