NFL Week 2 Picks: Every Game, Every Bet, September 20, 2026

SlateLines (Tally LLC) earns a commission when you sign up through our links. That never affects our analysis.

Fifteen games between Sunday at one and Monday night, a side on every one of them, and four with a full write-up. The method is the same in every row: our season win projections for the two teams, converted into points at the rate this week’s board itself pays for a win, three points for playing at home, and then a straight comparison against the number the books hung. No totals, no props.

The Whole Board

The conversion matters more than it sounds. Our August win projections have always been the input, and the rate used to turn a win into a point was half a point, a figure fitted to win-total disagreements rather than to team quality. Run against this week’s 16 spreads it made every large favorite look like a bad price, which is a bias and not an edge. Fitted properly the rate is one point per projected win with three for home field, and it accounts for 58 percent of what the board did with these games. Everything below is the leftover, and the leftover is the opinion.

Three of the fifteen carry a disagreement of four points or more, and our own rule treats that as a warning rather than a reason. Sorted by the size of our disagreement, our widest gaps have been our worst results, which is the opposite of how a real edge behaves. Those rows carry a side because the table is complete, not because we like them.

Featured Picks

• Green Bay at the Jets is the only staked play on the board. Our number is Green Bay by 2.3 and the line is three and a half, the market has come two points toward New York since it opened, and five Green Bay defensive linemen are on the injury report. A half play. Full breakdown.
• Indianapolis at Kansas City is a lean. The card rates these teams within two tenths of a win of each other, the board asks six after a half point came off toward Indianapolis, through a three touchdown win on one side and a three touchdown loss on the other. A half play at confidence two. Full breakdown.
• Carolina at Atlanta: the spread is a pass, and the bet is a Bijan Robinson same-game parlay. Our six point disagreement exists because our projection assumes a healthy Atlanta quarterback room, and Tua Tagovailoa has not practiced this week. The market moved four points on news our number cannot see. Full breakdown.
• The Giants at the Rams: the spread is a pass, and the bet is a Giants same-game parlay. Our projection and the board agree to a tenth of a point, the closest agreement on the whole board, and the line never moved when Myles Garrett went on injured reserve, which is why our side is the Giants and the points rather than the Rams. Full breakdown.

Where We Disagree With the Board

• Miami Dolphins at San Francisco 49ers: the board asks San Francisco 49ers by 13.5 points and our number says San Francisco 49ers by 6.4. That is 7.1 points of daylight. Take Miami +13.5.
• Carolina Panthers at Atlanta Falcons: the board asks Atlanta Falcons plus 2.5 points and our number says Atlanta Falcons by 3.8. That is 6.3 points of daylight. Take Atlanta +2.5.
• Cincinnati Bengals at Houston Texans: the board asks Houston Texans by 2.5 points and our number says Houston Texans by 7.8. That is 5.3 points of daylight. Take Houston -2.5.
• Philadelphia Eagles at Tennessee Titans: the board asks Tennessee Titans plus 7 points and our number says Philadelphia Eagles by 3.1. That is 3.9 points of daylight. Take Tennessee +7.
• Indianapolis Colts at Kansas City Chiefs: the board asks Kansas City Chiefs by 6 points and our number says Kansas City Chiefs by 2.9. That is 3.1 points of daylight. Take Indianapolis +6.
• Cleveland Browns at Tampa Bay Buccaneers: the board asks Tampa Bay Buccaneers by 8.5 points and our number says Tampa Bay Buccaneers by 5.3. That is 3.2 points of daylight. Take Cleveland +8.5.

Read those top rows as a list of games where something is wrong, and it is at least as likely to be our number as the price. Miami at San Francisco is the clearest case: no projection built in August should be handing out thirteen and a half points of opinion in September.

Bottom Line

One staked play and one lean on the spreads, plus a same-game parlay on each of the four games we wrote up. The Jets and the points is the half play, Indianapolis and the points is the lean, and the two national spreads are both passes for opposite reasons, so in those two games the bet is the in-game parlay in each article. That is a thin week by design: the number told us where it disagreed with us, we checked why, and in most cases the answer was that the market knew something we did not.

21+ · Gambling problem? Call or text 1-800-GAMBLER