When did the Cowboys play the Giants in Week 1?

Sunday 13 September 2026, 8:20 PM ET, at MetLife Stadium.

What was the spread for Cowboys vs Giants?

The Cowboys were 3-point favourites at the close, with a total of 47.5 points.

Who covered the spread in Cowboys vs Giants?

The Giants covered as 3-point underdogs: Cowboys 20, Giants 28.

Did Cowboys vs Giants go over or under?

Over: 48 points against a total of 47.5.

Who had the most passing yards in Cowboys vs Giants?

Jaxson Dart (NYG): 23/29, 230 yards, 3 TDs.

Who had the most rushing yards in Cowboys vs Giants?

Cam Skattebo (NYG): 18 carries, 81 yards, 1 TD.

Who had the most receiving yards in Cowboys vs Giants?

Isaiah Likely (NYG): 8 receptions, 78 yards, 2 TDs.

Who leads the Cowboys vs Giants series?

The Cowboys lead 14-3 over the 17 games since 2018.

What is the average score when the Cowboys play the Giants?

50.9 points a game: the Cowboys average 30.2 and the Giants 20.8 in their 17 meetings since 2018.

Cowboys at Giants Preview: Sunday Night Football Week 1 Picks, Odds and Projections

Published September 7, 2026 by Statz NFL.

The model makes Sunday night at MetLife a coin flip the market has priced as a Cowboys win. The under, the Giants lean, the scorers and the two published slips for Cowboys at Giants.

Two bad defences under the lights

Sunday Night Football, 13 September, 8:20pm Eastern at MetLife, which is 01:20 BST on Monday morning. Dallas have won five of the last six against the Giants, but the one they lost was the most recent, 34-17 at this stadium in Week 18 as three-point favourites, and the meeting before that was a 40-37 Cowboys win that cleared a 44.5 total by 32 points. The series averages 52.7 points a game and has gone over in four of the six. That is the context for a 48.5 total the model wants to go under, and I will explain why.

The 2025 tape is a study in imbalance. Dallas had the league's number one offence by yardage, 403 a game with 277.5 through the air, and the league's number 32 defence, 30.1 points allowed a game. They finished minus nine in turnovers. The Giants were mid-table on offence (22.4 points, 17th) with the fifth-best run game, and 26th in points allowed. The forward power rankings are brutal to both: Dallas 24th, New York 32nd and last. This is a game between two teams the model expects to be bad, and one of them is a two and a half point favourite on the road.

Where the game is won

The matchup grid produces two of the biggest edges you will see all season, one each way. Dallas running against a Giants run defence that ranked 31st is a 92% Cowboys edge. The Giants throwing against a Dallas pass defence that ranked dead last, 265.9 yards allowed a game, is a 92% Giants edge. Add Dak Prescott's passing game (first in the league) against the Giants' 17th-ranked pass defence and the Giants' run game against a Cowboys front that ranked 23rd, and every unit on the field has a favourable matchup. That usually means points, and the market agrees at 48.5.

The model does not, and the reason is the two offences it is projecting rather than the two defences. Dak Prescott projects 253 passing yards and 1.7 touchdowns, CeeDee Lamb 63 yards, and Javonte Williams only 10.6 carries for 48 yards behind him. Jaxson Dart projects 224 yards and 1.3 touchdowns with 38 rushing yards of his own, Malik Nabers 54 receiving yards on 6.4 targets. Those are 23-point offences, not 27-point ones, and the model has both at exactly that: 23.6 to 23.5.

The injury report leans Giants. Dallas are without safety Juanyeh Thomas and cornerback Josh Butler, with P.J. Locke questionable, which is three names in a secondary that was already the worst in the league. Cam Skattebo is questionable for New York, and his status decides the Giants backfield and one of the best scorer prices in the game.

The model's number

Cowboys 23.6, Giants 23.5. A coin flip: Dallas by a tenth of a point on the projection, by 0.9 on the model's margin, with a 47.2 total. The market has Dallas by two and a half, a 48.5 total, and the Cowboys at 1.68, which implies 59.5%. The model has them at 50.4%.

The bet is the under. At 1.98 on Kalshi, under 48.5 is a 57.5% call against 50.5% implied and it rates three stars. The model total is a point and a half lower than the line. The Giants plus two and a half at 2.17 on Polymarket is the loudest number in the game, a 59.9% model call against 46.1% implied, and it rates zero stars. Fourteen points of disagreement on a primetime spread is the model telling you it does not know something the market does, and I am not going to dress that up. It is a lean at a quarter point, and the moneyline at 2.53 is the same lean with more variance.

Under 48.5 points — Total, DAL @ NYG, Sun 8:20pm ET / Mon 01:20 BST — at 1.98 — Model total 47.2. Two 23-point offences on the projection, and a line that remembers last September's 77-point game. Game preview
Giants +2.5 v the Cowboys — Spread, DAL @ NYG — at 2.17 — A coin flip with a field goal of cushion, at plus money. Zero stars; the price says a quarter point is fine. Game preview

Scorers, props and the slips

The scorer board belongs to the Giants if Skattebo plays: 2.00 for a back the model has scoring 65 times in 100, the best raw price in the game, with the questionable tag attached. Javonte Williams at 1.58 is short for a back projected at 10.6 carries and 0.4 touchdowns, and the model has him under the price rather than over it. Malik Nabers at 2.45 is a 47% call, close to fair. Jaxson Dart at 3.00 to run one in is a 34% call with 38 projected rushing yards, and it is the one I would add at a quarter point.

The star-rated props are both unders on the receivers: Nabers under 62.5 receiving yards (projected 54.1, 58%) and George Pickens under 65.5 for Dallas (projected 59.2, 57%), each at 1.90 with just over two stars. Darnell Mooney over 10.5 receiving yards at 1.90 is the softest line on the board, with the model projecting 33.7 and clearing it 83 times in 100. Javonte Williams under 73.5 rushing yards is the same story as his scorer price: the model clears 73.5 only 16 times in 100 on a 48-yard projection.

The Single Game Parlays page has two slips here, and both are throwing scripts. Throwing Festival, at 6.86, is Dart over 217.5 passing yards, Prescott over 253.5 and Jake Ferguson over 27.5 receiving yards: the model makes it 19.8% against 14.6% implied. Second Receiver Ceiling, at 8.01, swaps Dart for Ferguson over 3.5 receptions and comes in at 19.9% against 12.5%. Notice that both slips want passing yards in a game where the single bet is the under. That is not a contradiction: the under is about points, the slips are about yards between the twenties, and a 23-23 game with two quarterbacks throwing 250 is exactly what the model projects. A quarter point on one of them, not both.

Darnell Mooney over 10.5 receiving yards — WR, NYG — at 1.90 — Projected 33.7 yards against a line set for a decoy. The softest number in the game. Game preview
Malik Nabers under 62.5 receiving yards — WR, NYG — at 1.90 — Projected 54.1 on 6.4 targets. The Giants run first, and the model has the volume spread wider than the line assumes. Game preview
Cam Skattebo anytime TD — RB, NYG (questionable) — at 2.00 — Only if he is active. A 65% model call at evens is the best scorer price on the night, and the tag is the whole risk. Game preview
Throwing Festival slip — Dart o217.5 + Prescott o253.5 + Ferguson o27.5 rec yds — at 6.86 — Three passing legs from one game state. Game preview

The verdict

A 24-23 game either way, and the under is the bet. Keep the Giants to a quarter point until the market tells you why it likes Dallas by more than the model does, and wait on Skattebo's status before you touch the scorer board. If he is out on Saturday, Tyrone Tracy Jr. inherits the role and a much better price than 2.00.

Put it to work

Game page: Cowboys at Giants Live lines, the model's edge table and the full prop board.
Single Game Parlays Both published slips.
Injuries Skattebo and the Dallas secondary as the week develops.