How the market priced NFL Week 4's Sunday and Monday games

Fifteen Week 4 games follow Thursday night, fourteen on Sunday, October 4 and one on Monday, October 5. Since the Week 3 Sunday games began, five of them have moved by five or more points of no-vig win probability at US-regulated sportsbooks, and one total has dropped seven points. Kalshi and Polymarket both quote every game, and their asks sit within two cents of the sportsbook consensus on all of them.

Book prices below were captured at 13:20 UTC on Sunday, October 4 and venue prices at 13:10 UTC, shortly before the first of the fifteen, Indianapolis at Washington, kicked off at 9:30 AM ET. Prices for the later games will keep moving before kickoff. Thursday's Pittsburgh at Cleveland game is not included.

The biggest moves since Week 3 began

Each comparison uses the same sportsbooks on both sides. For the starting point that meant three or four books, because that was all our line history held in late September. Spreads are from the home team's side, so a plus sign means the home team is the underdog.

• Arizona at New York Giants, spread -2.5 to +2.5, win probability 55.5% to 44.4% (-11.1 points)
• Green Bay at Tampa Bay, spread +1.5 to +3, win probability 48.6% to 40.1% (-8.5 points)
• Atlanta at New Orleans, spread -3.5 to -2.5, win probability 62.5% to 54.5% (-8.0 points)
• Detroit at Carolina, spread +3 to +3.5, win probability 43.2% to 36.4% (-6.8 points)
• Los Angeles Rams at Philadelphia, spread +2.5 to +3.5, win probability 43.7% to 36.9% (-6.8 points)
• New England at Buffalo, spread -5.5 to -7, win probability 69.1% to 73.6% (+4.5 points)
• Kansas City at Las Vegas, spread +5.5 to +4.5, win probability 31.7% to 33.8% (+2.1 points)
• Los Angeles Chargers at Seattle, spread -6.5 to -7, win probability 72.8% to 74.9% (+2.1 points)

The Giants went from 2.5-point favorites to 2.5-point underdogs. At our earliest capture, on September 7, two books had them as favorites of 6.5 to 7 points, at 72.6%. Tampa Bay's move came with a change in the total, which fell from 45.5 to 38.5 over the same span.

Two baselines are earlier than the rest. For Green Bay at Tampa Bay and Atlanta at New Orleans our history holds no capture with three or more books between Thursday night of Week 3 and Sunday night, so their starting point is September 25, 00:05 UTC, ten minutes before Atlanta at Green Bay kicked off.

The other seven games moved by two points or less of win probability. They are Dallas at Houston, Denver at San Francisco, Indianapolis at Washington, Jacksonville at Cincinnati, Miami at Minnesota, New York Jets at Chicago and Tennessee at Baltimore. Indianapolis at Washington, Jacksonville at Cincinnati and the Jets at Chicago show the same spread as they did before Week 3.

Two large moves are older than Week 3. Chicago was a 9-point favorite over the Jets at our September 7 capture, from two books, and is now a 3.5-point favorite, at 62.6%. Almost all of that happened before Week 3 began (61.1% at our baseline). Indianapolis at Washington was priced near 50-50 on September 7 and is now 33.2% for Washington, again with nearly all of the move before Week 3 (34.2% at baseline).

Totals

• Green Bay at Tampa Bay, total 45.5 to 38.5 (-7.0)
• Miami at Minnesota, total 43 to 38.5 (-4.5)
• Los Angeles Rams at Philadelphia, total 46.5 to 42.5 (-4.0)
• Atlanta at New Orleans, total 45.5 to 47.5 (+2.0)
• Kansas City at Las Vegas, total 45.5 to 47.5 (+2.0)
• Tennessee at Baltimore, total 44.5 to 42.5 (-2.0)

Green Bay at Tampa Bay and Miami at Minnesota share the lowest total on the board at 38.5. Jacksonville at Cincinnati and Detroit at Carolina share the highest at 51.5 across all seven books. Tennessee at Baltimore carries the largest spread, with Baltimore at -11.5, followed by Minnesota at -9.5.

Since Saturday morning

Most of the repricing is old. Between the 11:16 UTC capture on Saturday, October 3 and the 13:20 UTC capture on Sunday, four spreads changed. Green Bay at Tampa Bay went from Tampa Bay +3.5 to +3, and Tampa Bay's win probability from 38.2% to 40.2%, the largest move on the board over that span at 2.0 points. Indianapolis at Washington went from Washington +4 to +5, with Washington slipping from 34.6% to 33.2%, and its total fell from 47.5 to 46.5. Atlanta at New Orleans went from New Orleans -2.5 to -1.5, with New Orleans slipping from 54.9% to 54.1%. Miami at Minnesota went from Minnesota -10 to -9.5.

Those are the most common spreads across all seven books. The first table compares only the three or four baseline books, which is why it shows New Orleans at -2.5 where the full set shows -1.5.

Five other totals moved, each by a point or less. They are Green Bay at Tampa Bay (39.5 to 38.5), Denver at San Francisco (47.5 to 48.5), Dallas at Houston (48 to 48.5), Detroit at Carolina (51 to 51.5) and New York Jets at Chicago (43 to 43.5).

Kalshi and Polymarket against the books

Kalshi and Polymarket each list a winner market for all 15 games, so there are 60 venue prices, one for each team on each venue. All 60 passed the rule we apply to venue prices, described below, and none was dropped. The bid-ask spread was one cent on every one of them.

The gaps to the sportsbooks are small. Measured against the de-vigged consensus of ten US-regulated books, the median venue ask was 0.5 cents above consensus on both venues. Across all 60 prices the gap ran from 1.3 cents below to 1.7 cents above, and 41 of the 60 asks were above consensus. Because an ask includes the venue's spread and the consensus has the sportsbook margin removed, a gap of half a cent to one cent is the ordinary cost of buying at the ask. The largest gaps:

• Minnesota (Miami at Minnesota), Kalshi ask 83c, Polymarket ask 83c, book consensus 81.3%
• Chicago (New York Jets at Chicago), Kalshi ask 64c, Polymarket ask 64c, book consensus 62.7%
• Atlanta (Atlanta at New Orleans), Kalshi ask 47c, Polymarket ask 47c, book consensus 45.7%
• Seattle (Los Angeles Chargers at Seattle), Kalshi ask 76c, Polymarket ask 76c, book consensus 74.8%
• Arizona (Arizona at New York Giants), Kalshi ask 57c, Polymarket ask 56c, book consensus 55.7%
• Baltimore (Tennessee at Baltimore), Kalshi ask 86c, Polymarket ask 86c, book consensus 84.9%

Minnesota is the largest gap, at 1.7 cents. Both venues ask 83 cents, which is above the highest of the ten books (81.9%). The largest gap in the other direction is Kalshi's 36-cent ask for the Jets, 1.3 cents below consensus. The book and venue prices were captured ten minutes apart, so this is not a difference in timing.

Kalshi and Polymarket quoted the same ask, to the cent, on 21 of the 30 prices they share, and never differed by more than one cent.

The sportsbooks disagree with each other by more than the venues disagree with the books. Detroit at Carolina has the widest spread between books, 2.4 points of win probability. Carolina runs from 37.8% at DraftKings to 35.3% at Hard Rock Bet, and both venues ask 36 cents for Carolina, inside that range. Indianapolis at Washington is next at 2.1 points, with Washington between 32.7% and 34.8% and both venues asking 34 cents. Then come Kansas City at Las Vegas (2.0 points) and Green Bay at Tampa Bay (1.9).

Liquidity and what the asks do not tell you

Kalshi's winner markets have traded a median of about 813,000 contracts per game. Kansas City at Las Vegas is the busiest at about 3.12 million and Detroit at Carolina the quietest at about 181,000. Polymarket reports liquidity of between roughly $571,000 (Atlanta at New Orleans) and $1.72 million (New England at Buffalo) per game, with a median near $985,000.

Volume says how much has traded since listing, not how much can be bought at the ask right now, which is why we filter on the bid-ask spread and not on volume. Asks are before venue trading fees. Venue prices are captured about every half hour through the day and evening, and can be stale by the time you read them.

Injuries

The week's game designations are in. Our injury table, last updated at 10:01 UTC on Sunday, lists 131 designated players across the teams in these 15 games, of whom 68 are Out, 2 Doubtful and 61 Questionable. The Giants are the only team with none listed.

Three quarterbacks are listed Out: Baker Mayfield (Tampa Bay), Caleb Williams (Chicago) and Jayden Daniels (Washington). All three teams appear in the moves above. Williams and Daniels were also listed Out on the league's Week 3 report, and Mayfield was not. Beyond those weekly reports we hold no record of when a status changed, so we cannot say which came first, the designation or the price. We do not attribute any move in this note to an injury.

How the numbers were built

Line moves use the seven US-regulated books in Troxler's line history, which are DraftKings, FanDuel, BetMGM, Caesars, BetRivers, Fanatics and theScore Bet. Offshore books and Pinnacle are excluded. A starting point is the last capture before the first Week 3 Sunday kickoff (17:00 UTC on September 27) that held at least three of those books, and it is compared with the latest capture restricted to the same books. A game's line is the most common spread or total across the books in a capture. Win probabilities remove the vig by normalising each book's two moneyline prices to sum to one, then average across books. "Earliest capture" means the first price Troxler recorded, on September 7, from two books, and is not a book's opening number.

The consensus used to compare venues averages the de-vigged moneylines of every US-regulated book listing both sides, which was ten books for every game at the 13:20 UTC capture. A venue price counts only at its ask, only if both a bid and an ask are present, only if the ask is no more than five cents above the bid, and only if it was captured within the last three hours. Venues are priced at the ask here, and no midpoint is ever substituted.

This note describes where prices sit and how they have moved. It is not a recommendation. Every game and market is at troxler.io/games, and line history for each game is at troxler.io/line-moves.