NFL Poll of Polls: Week 4

Week 3 produced the season’s strangest Sunday yet. Atlanta went to Green Bay and won by three touchdowns, Chicago beat Philadelphia 27–7, New England lost by 29 in Jacksonville, and Washington knocked off the previously unbeaten Seahawks 33–31. The panel has responded.

For the first time this season there is a new No. 1, and it is unanimous: all eight outlets rate Buffalo first. Cincinnati fell seven places, Las Vegas climbed eight, and the Rams dropped six.

This is the same poll of polls as previous weeks — the week’s rankings standardised and pooled, with every team’s move charted.

Here is where the panel landed after Week 3.

Buffalo takes the top spot from Seattle, who drop to fourth after their first loss. San Francisco holds second, Kansas City climbs to third, and Minnesota is the week’s quiet riser, up five to eighth.

The panel was more agreed this week than last, not less: mean pairwise Spearman ρ = 0.937, up from 0.881 in Week 3.

This week’s matchups, priced

Every team plays in Week 4. All 16 games are priced below from the consensus alone, with a sandbox for building any matchup you like.

The panel

8 outlets this week. Mean distance from the consensus order, in rank positions:

There is no outlier this week. Every outlet sits within 2.12 positions of the consensus, and the agreement range (ρ 0.947–0.984) is the tightest of the season. Last week Sports Illustrated sat 4.2 positions out; this week Conor Orr’s column is back inside the group, and it is a proper 1–32 ballot again after last week’s list skipped a number and had to be dropped.

Sources and method

Every number above comes from one pipeline. Each outlet’s weekly 1–32 ballot is standardised, pooled in a measurement-error model, shrunk toward the panel mean, and published with a range reflecting how much the outlets actually disagree. The matchups are priced from the consensus score alone, using constants fitted on 6,719 completed regular-season games.

The method, the calibration and the week-by-week replication checklist live with the code rather than in every post:

• analysis/nfl-power-rankings/SOURCES.md — one section per outlet: URL pattern, cadence, markup and the specific traps. Follow this file to reproduce a week.
• analysis/nfl-power-rankings/METHODOLOGY.md — both derivations, and what the method does not claim.
• analysis/nfl-power-rankings/README.md — the pipeline and how to run it.

This week’s panel — eight ballots, every one verified complete:

Two regulars are missing, and neither is a parsing failure:

• NFL.com had not published a Week 4 column when this ran. The templated slug that worked for weeks 1–3 returns 404, the article is absent from NFL.com’s power-rankings hub, and Nick Shook’s author page still lists only weeks 1–3. On that evidence it is late rather than gone.
• Neil Reynolds’ NFL.com UK column also had not published. His column runs later in the week than the US desk’s, and in Week 3 it never appeared at all.

ESPN remains absent permanently. Every URL on the site — including the homepage — returns an empty HTTP 202 to a plain request, so it cannot be read unattended.

The panel agreed more tightly than in any previous week: mean pairwise Spearman ρ = 0.933, and all eight outlets put Buffalo first. That is what a clear consensus looks like — worth contrasting with Week 3, when the same pipeline found ρ = 0.881 and a genuine outlier.