The week 8 NFL schedule. Our ATS pick and our total pick on every game we called.
What am I looking at?
Every game on the 2026 week 8 NFL schedule with the latest captured line, the book that posted it, and our number beside the market. Every card carries our ATS pick and our total pick with the tier each one earned, or says No pick where our analyst panel did not agree on a side. 3 picks go free every week, released one at a time on Tuesday, Thursday and Saturday mornings and stamped with the moment they dropped; members see every pick. Where books have not posted yet, the slate says lines pending.
What the columns are, stated plainly
Six columns, the same on every card. Kickoff. The matchup with the market spread on each team and the market total with its book. Our number, which is the model's read and is context, not the call. The ATS pick and the total pick, both stated the same way with the tier they carry, the panel vote behind them. Then the Details link. A column with no call in it says No pick, because an absence is a result too and we would rather print it than leave a hole.
The line is the latest capture per game from our odds feed, book named on every row. Our number is the EL NFL Rating v3 difference plus 2.122 points of home field, shown when both teams carry a rating (0 at a neutral site: the London, Munich, Madrid, Paris, Rio, Mexico City and Melbourne games list a designated home team but nobody is at home, and the card says so). That home field figure is not a guess: it is the 2.122 intercept of the EL NFL Rating v3 second-stage margin model (docs/el-nfl-rating-v3.md), and the per-team home field variant was tested and rejected in that same document for making out of sample error worse. That number is context, never the call. The calls come from our analyst panel: three seats work every game and the synthesis is the ATS pick and the total pick printed on the card, each with the tier the panel earned and the vote behind it. A tier is how decisively the panel agreed, never a win probability. Where the market has moved off the number we took, the pick says both, ours first, so the reader knows which number is ours. Rest days come straight from the schedule and are free context on every row. 3 picks go free every week, released at 9:05am ET on Tuesday, Thursday and Saturday and public from that stamp on; every other pick and the line movement open with membership. Early in the week nothing has dropped yet, and the board says so rather than promoting a stand-in. Every card is the whole read. The Details link opens the full game page: the panel's written case, the matchup context, the line history, and every trend on the game.
Week 8 · 14 games · Thu, Oct 29 to Mon, Nov 2
Schedule: Edge Labs database, 2026 regular season. No lines captured yet for this week. Our number on 14 games. No calls on this board yet: the panel publishes on Tuesday. 14 games we did not call.
No game this week matches. Try the team's city or nickname.
Members see every pick, every week
3 picks here are released free every week, one each on Tuesday, Thursday and Saturday morning. Members see every card in full: the ATS pick and the total pick on every game the panel called, with the tier on each, the line movement from open to now, and every pick first, plus the public record behind every number. Founding access $24/mo.
Provenance: schedule from edgelabs.nfl_games (2026 regular season, loaded by scripts/el_nfl_games_load.py from nflverse schedules); the date column is the schedule's gameday and gametime (Eastern) from the nflverse schedule. Lines from edgelabs.nfl_lines, latest capture per game, book named on every row. Our number is the EL NFL Rating v3 (edgelabs.el_nfl_rating, season 2026, model_version v3, 32 rows, docs/el-nfl-rating-v3.md) difference plus 2.122 points of home field, which is the 2.122 intercept of the EL NFL Rating v3 second-stage margin model (docs/el-nfl-rating-v3.md), and it is context on the card, not the call. The ATS pick and the total pick come from the analyst panel (edgelabs.nfl_sme_picks, migration 0091), three seats per game synthesized into one call with a tier; the ATS call is stamped into edgelabs.pick_locks at the number we took, and nothing here is a win probability. Line movement comes from our permanent capture log (edgelabs.nfl_lines_history), one book per game across both ends and never a capture taken at or after kickoff, so the second number is the last pregame line; a game the log has captured only once shows no movement rather than an invented one. Rest days are the schedule's own home_rest and away_rest. The v3 document states plainly that the rating is a margin and expected wins engine, not a proven spread beater; treat every call here as research, never a guarantee. 21+.