The Dread
Sheet

Every score, and how last week's picks did against them: 3-2 for the week, 16-9 on the season. The 74.9% and 69.9% picks both missed and the 52.5% coin flip won by 54.72 — 25 picks is too few to say whether the confidence numbers mean anything yet.

PA doubles as schedule luck, all-play is the record against all nine teams every week, and Wk 5 bench is what the best possible lineup would have added. H00dedWizard (42.2), No. 5 (17.4) and Smallwood Retrievers (17.1) each benched more than they lost by. Team Fawley and winniehartley share the best all-play, 33-12; NoseyNabers and OldeRowanRiot share the worst, 10-35. jfinau has faced 763.4 points, the most in the league; No. 5 has the softest schedule, 687.8.

Top Performers

Best actual starts, and the bench decisions that'll sting longest.

Three of the five bench entries are quarterbacks, and two of those cost something: Daniels and Stafford out-scored the starters. Lawrence's 22.7 did not — Joe Burrow started and scored 24.6. Last week's start/sit and free-agent calls, scored: Bucky Irving for winniehartley (31.5), Rhamondre Stevenson for No. 5 (17.5), Justin Jefferson over Luther Burden for trevorh84 (18.8 against 8.9), and the free-agent adds of Jacoby Brissett (26.3) and Emanuel Wilson (13.2) were right and followed. Isaiah Likely (32.4), Ollie Gordon (14.8) and Terry McLaurin (12.2) were right and ignored. RJ Harvey over Juwan Johnson (0.7 against 5.6) was wrong and followed. Bo Nix over Jordan Love (7.9 against 25.3), DeVonta Smith over Keon Coleman (0.0 against 7.7) and Rico Dowdle over Omarion Hampton (0.0 against 4.9) were wrong and, luckily, ignored. And Ja'Marr Chase, whom this page called Out, played.

Every move since the last blotter ended on Oct 4, pulled across both transaction rounds so nothing between editions slips through. Wednesday's waivers cleared at 3:05; after that it's free agency, first click wins. Green is a player arriving, red a player leaving. Times are Eastern.

The real 14-week schedule played out 50,000 times. Each team's weekly mean is a 50/50 blend of its optimal Week 6 projection and what it has actually scored, with a 27-point standard deviation; top six make it. Same method as last week, so every change below is the week, not the model.

H00dedWizard fell 84.6% to 41.5% on a 0.37-point loss and the league's lowest Week 6 projection, and drops out of the top six. Unauthorized Use of No. 5 fell 96.8% to 83.3%. She'sTheMann rose 81.3% to 98.4%, OldeRowanRiot 3.1% to 17.5% and trevorh84 65.9% to 78.7%. Two of those climbs lean on players listed Out: OldeRowanRiot's projection counts 24.8 from Jalen Coker and DeVonta Smith, She'sTheMann's 20.8 from Lamar Jackson, and half of each team's mean is that projection. She'sTheMann's 25.2% title chance is the best on the board, with the top three within 4.4 points of each other. Smallwood Retrievers has a 47.6% chance of finishing last. Nine games remain.

Five predictions for Week 6, timestamped before kickoff so they can be wrong in public again. Each side's number is its best legal lineup, not what its manager will actually start — Team by Team, below, has that gap.

The flattest slate of the season: no margin above 4.5 points, no pick above 54.7%. Four teams are on bye, and several numbers still count players listed Out — Lamar Jackson for She'sTheMann, Jalen Coker and DeVonta Smith for OldeRowanRiot, Breece Hall for Team Fawley, Saquon Barkley for No. 5. Zeroing them moves every margin (G3 to 5.1, G4 to 2.3, G5 to 12.2) but flips none. They stay as Sleeper has them, because last week this page called Ja'Marr Chase Out on Sleeper's word and he scored 19.7. Come back Tuesday and find out how this one aged.

Ranked on each team's best legal Week 6 lineup, which is why the order differs from the odds table. The bench gap is what that lineup beats the one actually set, if nothing changes before kickoff. Cincinnati, Detroit, Miami and Minnesota are on bye: fourteen starters on six teams, and five teams have nobody healthy behind one of them.

Every unrostered player with a real projection, tried against every rostered player on each team, through the same optimal-lineup math as everything else on this page. One suggestion per team: whichever drop-for-add raises that team's best-lineup total the most.

Six teams want Woody Marks and two want Trey Smack — first click wins. Several of these fill a bye-week hole rather than upgrade anyone: the model drops Jake Bates, Sam LaPorta and Will Reichard because they score zero this week, not because they're done. H00dedWizard needs both a running back and a kicker; this table only shows it one. Jalen Hurts, dropped Friday by winniehartley, is also unclaimed.

Every pair of rosters, run through the same optimal-lineup math: does swapping these two players raise both sides' best-lineup total? Ranked by whichever team benefits less — the point is mutual upside, not a heist.

The top pair leans on Breece Hall, who is listed Out, so H00dedWizard's side of it is softer than it looks. A lot of this list is the bye week talking: Eddy Pineiro shows up three times because three teams' only kicker is off this week.