The Rams lost at home to Buffalo and Arizona and are 2-4, and our model has cut Los Angeles to ten wins with the title odds at 8.5 percent
Updated
Los Angeles lost 37-34 at home to Buffalo as a 92 percent favorite, then lost 18-16 at home to Arizona as a 98 percent favorite, and the Rams are 2-4 and fourth in the NFC West on the tiebreaker. When we ran the rest of the season they came out at 10-7 with a 72 percent playoff chance and an 8.5 percent title chance. Two weeks ago those numbers were 13-4, 99 percent and 29 percent. In July they were 14.8 wins and 44 percent. The team our model called the best in football for four months is now the fourth-most likely champion, level with Jacksonville, and that is the largest fall any team has taken in a single run this season.
Two things produced it. The first is the football. Matthew Stafford is 164 of 249 for 1,794 yards with eleven touchdowns and six interceptions, a passer rating of 92, and the defense that projected second in points allowed in July has given up 55 points in two home games to Josh Allen and Jacoby Brissett. The second is that our model now leans harder on this season's snaps than it did a month ago, which means four losses in six weigh more than they would have in September. Stafford projects for 4,831 yards and 30 touchdowns against 15 interceptions behind a 94 rating, down from 100 in the summer. Puka Nacua is back to full volume and projects for 127 catches and 1,544 yards. Kyren Williams projects for 1,058 rushing yards and ten touchdowns, down from 1,354 and 13. Konata Mumpfield's projection has jumped from seven catches to 41 as the rookie has taken on a role. The offense projects for 394 points, twelfth in the league, down from 454 two weeks ago and 551 in July.
The defense is still graded as a top-six unit, which is the model telling you it has not given up on the roster. Los Angeles projects to allow 320 points, sixth in the league, up from 280 and second two weeks ago. Byron Young projects for 12.3 sacks. Aaron Donald for under seven. Quentin Lake's interception projection has climbed from two and a half to nearly four and his tackle line from 81 to 96, which is the sign of a safety making a lot of stops after the catch. Kamren Curl projects for three and a half picks. The team projects for 39 sacks, 20th, and nearly 15 interceptions, fifth. The unit has allowed 138 points in six games, 23 a game, and a 320-point projection is 18.8 a game from here. The model still expects the defense to be better than it has been. It just expects it less confidently.
The division is gone as a formality and the model is honest about it. San Francisco is 6-0 and projects for twelve wins with the head-to-head banked. Seattle is 3-3 and projects for seven. Arizona is 2-4, projects for six, and holds the tiebreaker over the Rams after Sunday. Los Angeles is four games behind the 49ers with eleven to play, and when we ran the rest of the season the Rams won the NFC West in a minority of simulated years for the first time all season. The 72 percent playoff chance is now mostly a wild card number.
The next month is the softest stretch of the schedule and it arrives at the right time. A trip to Las Vegas at 76 percent, the Chargers at home at 88, a trip to Washington at 60, and a trip to Arizona at 82. When we ran that stretch Los Angeles went 3-1 more often than any other outcome and 4-0 nearly as often, and a 6-4 record entering the second week of November would put the ten-win projection on schedule and the Rams back in the thick of the wild card race. The model expected the Rams to beat Arizona last week too, which is the caveat.
The two home games were the whole story and they were different failures. Against Buffalo the Rams scored 34 and lost, because the defense gave up 37 to a Bills offense that has now scored 30 or more in four of six games, and the model's second-ranked points-allowed projection took its first real hit. Against Arizona the Rams scored 16 and lost to a Cardinals team that had given up 45 to Detroit the week before, and that was the offense failing against a defense the model has 31st. Two different halves of the roster lost two different games in two weeks, and that is why the projection moved three wins rather than one. Stafford's line tells the story in miniature: his completion percentage is fine, his yardage is fine, and six interceptions in six games against a projection that had him at 14 for the season is the number the model has reweighted hardest. The offensive points projection has dropped 60 in two weeks. What moves the number back is four games the model has the Rams winning comfortably, starting in Las Vegas. If Los Angeles is 6-4 in mid-November the ten-win projection will have been right and the title odds will have crept back up. If it drops one of the four, the model will do to the Rams what it did to Seattle.
The matchup record is the worst in the league against the projections and it is not close. We gave Los Angeles 90 percent against San Francisco, 96 against the Giants, 80 in Denver, 70 in Philadelphia, 92 against Buffalo and 98 against Arizona. That is more than five and a quarter expected wins from six games, and the Rams have delivered two. Only the Chargers, at 0-6, have been further below their expected wins. The market opened Los Angeles at eleven and a half wins. In July we sat more than three over it, which was the largest lean on the board. We now sit a win and a half under it. The roster the model loved is still mostly on the field. It has lost four of six, three of them at home, and the model has priced six weeks of that at what six weeks is worth.
