Commanders at Cowboys Final Score: Dallas Pulls Away and the Over Cashes

Dallas won 37-20. We had two positions here and split them: Washington +4 lost by 13, and the over 51 won with 57 points scored.

The spread call was built on Dallas having won by more than four at home once last season. They won by 17. That argument was about a pattern and the pattern broke on the first Sunday it was tested, which is what happens when one season is the whole sample.

Dallas scored in every quarter: 10, 7, 10, 10. Washington never led. They out-gained Dallas 389 to 332 and converted 8 of 14 on third down, and still lost by 17, because Dallas went 3 for 4 in the red zone and Washington went 1 for 3.

The over was never in doubt. 17 points in the first quarter does most of the work on a 51 number.

The number

Dallas closed at -4 and covered by 13. The total closed at 51 and the game went 57, so the over won by six.

What we said before kickoff

The prediction below is unchanged, so you can see what was argued and judge it.

Dallas won by more than four at home once last season. Take Washington +4 and the over.

This page took Washington +4.5 and the over 50.5. The over has not moved: 50.5 at all eleven books, read on 17 September 2026. The spread has come in half a point. Washington is +4 at nine of the eleven books, DraftKings still has +4.5 at -115, and FanDuel has +3.5 at -105.

Source: US sportsbooks via Nucleus, read 2026-09-17 05:24 UTC. Each point is the newest time any book quoted that line (times UTC).

The last time any book quoted Washington +3 was 14 September. So the market has moved a point toward Dallas, and I am still on the other side of it.

My bias, up front: I like the underdog in a division game. Dallas beat Washington twice last season, and I will come to both games.

Pick: Washington Commanders +4 (-110)

Dallas is laying four at home. So how good was Dallas at home last season?

Not good enough to lay four. The Cowboys went 4-3-1 in their own building in 2025, scoring 29.9 points a game and allowing 30.4. A team that was outscored at home is asking you to believe it wins this one by five or more.

Source: NFL 2025 results via Nucleus, checked game by game against ESPN. Neutral-site games are left out of both.

Look at the eight home margins. Dallas beat the Giants 40-37, Philadelphia 24-21 and Kansas City 31-28. It tied Green Bay 40-40. It lost 27-17 to Arizona, 34-26 to Minnesota and 34-17 to the Chargers. Washington +4 wins the bet in seven of those eight games.

The eighth is the problem, and I will say it plainly. Dallas beat Washington 44-22 at home on 19 October 2025, the only time it won by more than four at home all season.

Source: NFL 2025 results via Nucleus, checked game by game against ESPN.

Washington’s road record is the other doubt. The Commanders went 3-5 in true road games in 2025, scoring 19.2 a game and allowing 26.5, and all five road losses were by more than four: 27-18 at Green Bay, 34-27 at Atlanta, 44-22 at Dallas, 28-7 at Kansas City and 31-0 at Minnesota.

Over the full season neither team earned much trust. Dallas finished 7-9-1 and was outscored by 40 points. Washington finished 5-12, was outscored by 95, and went 2-5 in games decided by eight points or fewer, against 4-3-1 for Dallas.

I overrule all of it, and here is why. A 22-point win is one game. In seven of eight home games Dallas failed to win by more than four, and that habit is exactly what a four-point spread prices. I would rather back the habit than the blowout.

What Week 1 said

Both teams lost on the road in Week 1. Only Dallas was favoured to win.

Washington lost 24-22 at Philadelphia as a 6-point underdog, covering by four. It trailed 17-9 after three quarters and scored 13 in the fourth.

Source: ESPN scoreboard, read 16 September 2026; the period scores add up to both finals.

Dallas lost 28-20 at the New York Giants as a 3-point favourite and missed the cover by 11. The Cowboys trailed at the end of every quarter: 7-0, 14-7, 21-14 and 28-20.

One game each proves nothing on its own. It does not argue for the move from +3 to +4 either.

The division

This is an NFC East game, and last season the division went Dallas’s way. The Cowboys were 4-2 inside it and Washington was 3-3.

Dallas also won both meetings. The second was closer than the first: 30-23 as the visitor on 25 December 2025, a seven-point game.

Both of those results would have beaten Washington +4. Both also went over 50.5, with 66 points in October and 53 on Christmas Day, which is the reason the second bet exists.

What the moneyline says

The median of eleven books has Dallas at -208 and Washington at +175. Take the bookmaker’s margin out and that prices Washington to win 35.0% of the time.

Source: US sportsbooks via Nucleus, read 2026-09-17 05:24 UTC.

At +175 the bet needs 36.4% to break even. The market’s own number sits below that, so I leave the moneyline alone and take the points.

Price discipline on the spread: +4 runs from -105 at LowVig to -112 at BetRivers, a break-even of 51.2% to 52.8%. DraftKings’ +4.5 at -115 needs 53.5% and turns a four-point Dallas win into a winner. FanDuel’s +3.5 at -105 is still a bet. At +3 I pass, because three of Dallas’s home wins last season were by exactly 3.

The other side is no cheaper. Dallas -4 runs from -105 at LowVig to -111 at Caesars, 51.2% to 52.6%, and FanDuel charges -115 for Dallas -3.5, 53.5%. With both sides priced alike, the choice comes down to those eight home games.

Pick: Over 50.5 (-108)

The total has come down two points without dropping below where I want it. It was 52.5 on 9 September, 51.5 on 14 September, 49.5 later that day and 50 on 16 September. It is 50.5 at every book now.

Source: US sportsbooks via Nucleus, read 2026-09-17 05:24 UTC. Each point is the newest time any book quoted that line (times UTC).

Dallas games go over this number. Twelve of the Cowboys’ 17 games last season finished above 50.5, with a median of 57. At home it was six of eight: 77, 80, 66, 59, 60 and 51.

Source: NFL 2025 results via Nucleus, checked game by game against ESPN. The line: median of the books quoting the game.

The doubt is Washington on the road. Its eight true road games had a median of 43, and six of them stayed under 50.5. Put those beside Dallas’s home games and the count is eight over, eight under.

I take Dallas’s side of that tie. Dallas home games averaged 60.3 points between the two teams, Washington road games 45.8, and halfway is 53.0, above this number. The Cowboys allowed 30.1 points a game over the season, and both of their meetings with Washington went over. Both Week 1 games also went over their closing totals: Washington’s by two points, 46 against 44, and Dallas’s by half a point, 48 against 47.5.

The over price runs from -104 at LowVig, a break-even of 51.0%, to -114 at BetRivers, which needs 53.3%. The under costs -108 to -115, 51.9% to 53.5%, so the price does not push me off the over. It is a bet at 50.5 and at 51. At 51.5 I pass.

What changes it

A line of Washington +3. Three of Dallas’s eight home results last season were three-point wins, and at +3 each of those is a push.

A total of 51.5 or higher. I want the over at the number the whole market has now, and not at the one it moved away from on 14 September.

Read both injury reports before you bet. Nothing in this piece says who is available, and a spread that drops to +3 before Sunday is the first sign something has changed.

Washington Commanders at Dallas Cowboys, Sunday 20 September 2026, 4:25 PM ET. Prices: US sportsbooks via Nucleus, read 17 September 2026, 05:24 UTC. 2025 results: Nucleus, checked game by game against ESPN. Week 1 period scores: ESPN. Every Week 2 game: NFL Week 2 odds and predictions.