Last Updated on Novembro 17, 2026 7:28 pm by Erwin Noguera
Last Tuesday we closed this NFL column with a prediction about the market: “sides are being priced accurately, and totals are not. If there is value left on this board, it is on the over-unders.”
It was exactly backwards.
The average miss on the totals fell from 10.2 points to 8.36. Favorites went from 14-1 straight up to 10-4, and from 11-4 against the spread to 9-5. One underdog won outright in Week 9. Four won in Week 10.
The totals got sharper, and the sides fell apart, and we told you to expect the opposite.
One of those four underdogs was 0-8. And the highest total on the board went thirty-and-a-half points over.
Uma nova era de As apostas esportivas ponto a ponto chegaram pela TSMx.. Negocie apostas esportivas ponto a ponto para obter melhores odds e ganhe 1.000 TSMx Coins + $5 em Crédito TSMx ao fazer isso. Cadastre-se agora mesmo através do nosso link especial..
Week 10, Against the Numbers
| Jogo | Resultado | Total | Linha | vs total | ATS |
|---|---|---|---|---|---|
| San Francisco at Dallas | 46-40 | 86 | 55.5 | +30.5 | SF -3.5 covered |
| Jacksonville at Tennessee | 40-23 | 63 | 45.5 | +17.5 | JAX -10.5 covered |
| Houston at Cleveland | 24-27 | 51 | 40.5 | +10.5 | HOU -4.5 lost |
| Miami at Indianapolis | 22-24 | 46 | 37.5 | +8.5 | IND -5.5 lost |
| Pittsburgh at Cincinnati | 18-38 | 56 | 49.5 | +6.5 | CIN -4.5 covered |
| Washington at NY Giants | 24-21 | 45 | 40.5 | +4.5 | NYG -1.5 lost |
| LA Chargers at Baltimore | 23-20 | 43 | 39.5 | +3.5 | BAL -3.5 lost |
| LA Rams at Arizona | 34-6 | 40 | 38.5 | +1.5 | LAR -8.5 covered |
| Kansas City at Atlanta | 35-13 | 48 | 47.5 | +0.5 | KC -3 covered |
| Minnesota at Green Bay | 30-10 | 40 | 42.5 | -2.5 | GB -2.5 lost |
| New England at Detroit | 10-35 | 45 | 50.5 | -5.5 | DET -2.5 covered |
| Seattle at Las Vegas | 21-13 | 34 | 40.5 | -6.5 | SEA -6.5 covered |
| Buffalo at NY Jets | 31-12 | 43 | 50.5 | -7.5 | BUF -8.5 covered |
| Carolina em Nova Orleans | 20-17 | 37 | 48.5 | -11.5 | CAR -1.5 covered |
Nine overs, five unders, and only one game missed its number by more than twenty: against four the week before.
Three games finished within three points of the total and three within three points of the spread. Kansas City at Atlanta landed half a point over 47.5 and twenty-two points past the line, which is the week in miniature: the total was right, and the side was nowhere near.
Vencedores
San Francisco, in the game of the season so far. 584 yards, 46 points, and not a single turnover in a 46-40 win at Dallas. Brock Purdy went 22 of 28 for 389 yards and three touchdowns: 78.6% completion at 13.89 yards per attempt. Christian McCaffrey caught four passes for 161 yards and two scores, 40.25 yards per catch. Deebo Samuel carried three times for 78. They are 8-1.
Cincinnati, against the fourth-best defense in football. Joe Burrow went 35 of 44 for 409 yards and four touchdowns, 79.5% completion, in a 38-18 win over Pittsburgh. Ja’Marr Chase caught eleven for 231. 484 yards, thirty-first downs against Pittsburgh’s seventeen, and no turnovers. This was the game our four-point rule told us to take, and it was not close.
Jacksonville, by simply keeping the ball. 38 minutes, 10 seconds of possession against 21:50: a sixteen-minute advantage on the way to 40-23 at Tennessee. 459 yards, twenty-seven first downs, Trevor Lawrence 25 of 35 for 352 and three touchdowns, Parker Washington seven catches for 152. Their differential is now plus 125, thirty-seven clear of second place.
Buffalo, on the ground. James Cook carried eighteen times for 157 yards (8.72 yards per carry) in a 31-12 win at the Jets. Josh Allen threw the ball twenty-one times. When the run game produces at that rate, Buffalo does not need him to.
Minnesota, which nobody expected to win. Thirty points on the road as 2.5-point underdogs, from the twenty-fifth-ranked offence in football. Green Bay turned the ball over four times and were sacked eight times for 59 yards. Kirk Cousins went 21 of 31 for 207 and two scores.
And the Chargers, finally. 23-20 at Baltimore for their first win of the season, and they did it on 287 total yards, 129 passing yards and two conversions from thirteen third downs. Zero turnovers. 160 rushing yards against Baltimore’s 16. Omarion Hampton carried twenty-two times for 105. It was not pretty, and it does not need to be.
Perdedores
Baltimore, at home, to a winless team. Derrick Henry carried ten times for ten yards. The team rushed for 16 yards in total. Jacoby Brissett went 33 of 49 for 340 yards and no touchdowns; they turned it over twice, and they are 4-6 with a minus-50 differential. They outgained the Chargers 347 to 287 and lost.
Dallas, who scored forty points and lost. Dak Prescott threw five touchdowns. George Pickens caught eleven for 145 and three of them. They gained 439 yards. And they are 4-6, because they concede 28.20 a game, thirtieth in the league, while scoring 28.60, fifth. A plus-four differential across ten games is what that combination produces.
Baía Verde. 220 yards, four turnovers, eight sacks conceded, beaten 30-10 at home by a side averaging 19.56 points. Jordan Love went 17 of 27 for 220 with an interception, a fortnight after we watched him throw for 372 and called it a one-week event. They are 6-4.
Pittsburgh, whose defense was the whole argument for them. They came in fourth in points allowed and gave up 484 yards and 38 points. Their offence managed 222 yards and 120 through the air, with Aaron Rodgers 15 of 29 for 143. Payton Wilson made nineteen tackles, which tells you how much football was played in front of him.
Arizona scored six points. Carson Beck went 11 of 18 for 108 yards with two interceptions in a 34-6 defeat, and Gardner Minshew’s season has since been ended by a torn UCL in his throwing arm. They are 3-7 with a minus-105 differential, the worst in football.
And Houston beat themselves. Thirteen penalties for 112 yards and three conversions from twelve third downs in a 27-24 loss at Cleveland as 4.5-point favorites. Cleveland gained 379 yards to Houston’s 290. The Texans are 3-6.
Reações exageradas
“The Chargers have turned it around.” They are 1-8 with a minus-95 differential, and they won with 287 yards and a 15.4% third-down rate. One win built on the opponent rushing for sixteen yards is not a turnaround. But (and this is the part we have to say) ESPN’s FPI called this game and we did not.
“Dallas should start over.” They are fifth in the NFL in scoring. The problem is a defense ranked thirtieth, and a defense is half of a team you can rebuild in one offseason without touching the quarterback. A 4-6 record from a top five offence is a diagnosis, not a demolition order.
“Minnesota’s offence is fixed.” They scored thirty once, against four turnovers and eight sacks. They average 19.56 points a game and rank twenty-fifth. The defense, at 14.00 points allowed, is still two full points better than anybody else in football and is still carrying the entire team.
“Jacksonville is the best team in the league.” By point differential, they are, and it is not close at plus 125 against Buffalo’s plus 88. They are also 7-2 behind two 8-1 clubs, and a differential built partly on 40-23 at a 1-8 Tennessee is not the same as a differential built against the schedule Buffalo have played. They are a genuine contender. They are not yet demonstrably the best.
“Green Bay is finished.” They are 6-4, and their differential is plus five. That is a .600 team with essentially no margin, which is exactly what it was before Sunday. One four-turnover afternoon does not change a team that was already mediocre on the underlying numbers.
Onde nos enganamos
The closing paragraph of last week’s column, and we are putting it at the top of this section rather than the bottom.
We wrote that the market was pricing sides accurately and totals badly, and that the value left on the Week 10 board was on the over-unders. The totals then improved by 1.84 points of average accuracy while favorites lost four games outright and five against the number. If you followed that read, you were looking in the wrong half of the board all weekend.
The lesson is not that we had the direction wrong. It is that one week of total misses is not a signal. A 10.2-point average across fifteen games is a small sample dressed up as a finding, and we presented it as a finding.
Our four game predictions all landed: Jacksonville, Buffalo, San Francisco, and Cincinnati, every winner and every spread, with Buffalo’s 31 exactly right. That is the second week running we have gone four from four, and it is worth considerably less than it sounds given what the rest of the board did.
The Chargers’ call is the one that stings, and we marked it in advance. We wrote that the 21.3-point disagreement was “pointing at a winless team,” that we were not betting it, and that we were “also not going to pretend the model is obviously wrong when it has been right more often than we have.” It was right again. Of the six games we named as past our four-point threshold, the model went 5-1, and the only one of the six we declined to follow is the one it won most emphatically.
O único número que realmente importa
8.36 points. That is the average distance between the closing total and the actual score across all fourteen games, down from 10.2 in Week 9.
And the spread numbers moved the other way. Favorites 10-4 straight up against 14-1. Nine and five against the number against eleven and four. Four outright underdog winners against one.
Put those together and the honest conclusion is that neither figure was ever telling us much. Two weeks, two opposite readings, from the same market. What actually happened is that one week had a 72-point game and a 66-point game in it, and the next week had an 86-point game and not much else.
The thing that did hold was the model. 5-1 on the games we flagged in Week 9, and the rule we wrote down on November 3 has now been tested twice and passed twice. Our own record on it is worse than the models, because we have overridden it once and the override cost us the biggest call of the weekend.
The practical read for Week 11: ten of thirteen games clear the four-point threshold, up from seven of fourteen. That is the model disagreeing with the market on 77% of the board. We are going to follow it where it points and stop building theories about which half of the board is mispriced.
No Gambyl, você pode acompanhar toda a temporada da NFL. Confira as últimas notícias, previsões e probabilidades de apostas enquanto desfruta de uma experiência completa de cassino online em Cassino Gambyl, Com ação rápida e jogabilidade intuitiva.