Week 6 Regression Can Be Mean

October 13, 2026

After 5 weeks, we’re getting our lay of the land.

For some of us, that means a chance to right the ship, if things have started rough.

Either way, we’ve got 5 full weeks of data collected now, which means we’re also better able to evaluate what moves will give us best chances. We can usually count on planning our next move with a little more predictability.

However, it’s also a week with 4 Byes. No Bengals or Lions to plug in on our offenses. No Vikings as our Defenses. And no Dolphins to stream against!

Still, I see plenty of good streaming options out there, and I hope you feel armed to make your decisions with all the “Details” you need.

Reminder of a couple updates

1. Kicker “Hand-Picked” Rankings. Introduced a couple weeks ago, you can view Kickers in my curated ordering. They’re not updated more than daily, but the hope is to help you with decisions where kickers are very close in projection.

2. Snapshot FLEX view. The FLEX positions now appear on the Snapshot view— as 3 simple lists by position (RB / WR /TE). The Details page still lets you customize your view a bit more, and soon we hope to add some customization for scoring settings as well.

Week 5 In Review

Last week, the front page contained the Accuracy Report for weeks 1-4.

Meanwhile, week 5 was not a week that lived up to the excellent promise of that report.

The biggest upset of all was the Cowboys-Buccaneers game on Thursday. It killed the first Survivor pathway, and the Betting Lines, and it killed D/ST accuracy as the Cowboys went negative in ESPN scoring. The Jaguars-Eagles game in London was almost as upsetting, from an accuracy standpoint.

• D/ST: All rankings sources had a tougher week, where both Jaguars and Cowboys were among top options. Subvertadown was just “average” for the week.

• Kicker: It was actually a great week for experts being able to predict kickers— Subvertadown included. A lot of rankers failed with the Mevis pick (including me), but otherwise no complaints. I noted only 2 sources outperform me for the week.

• QB: Quarterbacks were exceptionally predictably for the week. Everyone was wrong on Dak Prescott, and I was wrong on Bo Nix. But otherwise what a good week— my models had the lowest Accuracy Gap.

• Betting Lines: The biggest disappointment of the week. It was the worst week in 2 years for the models, as the -40% ROI wiped out the great +33% gains made in week 4 just before. I wrote just last week that things would swing low again, but did not expect such a blow to happen so soon— and not with the Cowboys! For the season, Betting Lines are still slightly net positive, but it will take some rebuilding now.

• FLEX positions: WR and TE went really great for my models, again! At RB, I was slightly worse than the average, by having Bijan and Swift too high.

Let’s do Week 6!

No other news as we plough ahead.

Just remember we’re still waiting word on injuries and returning players, and these get updated throughout the week.

You can always use the Injury Inquiry tool (on the D/ST page) to ask about any inclusions / exclusions you’re curious about, for the effect on the rankings.

Good luck!

/Subvertadown