Edge
Build your Edge
Where we differ from the market. Build a model, test it on graded games, or start from the Lab, your models, or a remix.
Where we differ from the market
Lines 10:57 PM PDT · 9 plays
Play board · ordered by model confidence, highest first · ties by kickoff
9 house plays · Week 6 · Thu, Oct 15 – Mon, Oct 19
9 with a captured price
| # | Game | Play | Model | Best price |
|---|---|---|---|---|
| 1 | NYJ @ NE | NE −7 | confidence 63/100 | −101 ProphetX at −7 captured Oct 13, 1:57 AM ET |
| 2 | DAL @ GB | GB −1 | confidence 63/100 | −102 ProphetX at −1 captured Oct 13, 1:57 AM ET |
| 3 | WAS @ SF | SF −7.5 | confidence 63/100 | +102 Novig at −7.5 captured Oct 13, 1:57 AM ET |
| 4 | CAR @ PHI | PHI −1.5 | confidence 60/100 | +108 Kalshi at −1.5 captured Oct 13, 1:57 AM ET |
| 5 | CHI @ ATL | ATL +3 | confidence 59/100 | +108 LowVig at +3 captured Oct 13, 1:57 AM ET |
| 6 | SEA @ DEN | DEN +3 | confidence 58/100 | −106 Livescorebet at +3 captured Oct 13, 1:57 AM ET |
| 7 | TEN @ IND | IND −7 | confidence 57/100 | +102 ProphetX at −7 captured Oct 13, 1:57 AM ET |
| 8 | PIT @ TB | TB +3 | confidence 57/100 | −112 ProphetX at +3 captured Oct 13, 1:57 AM ET |
| 9 | BAL @ CLE | CLE +4.5 | confidence 55/100 | +100 Novig at +4.5 captured Oct 13, 1:57 AM ET |
Swipe the table for model · best price
Order = the model number printed in each row, highest first. A tie goes to the earlier kickoff. A spread or total shows a confidence out of 100. It ranks plays. It is not a win chance. Best price = the best captured book price at the play's own line. A missing capture reads as its words, never as −110.
Sharp money, latest
No spread moves to show for this slate yet. They appear once the books post ticket and money shares.
Where to go next
- Lab — ask the question first
Pick a catalogue question and read its sample, baseline and lift before you build anything. Answers are precomputed, so the number you see is the number that was graded.
152 answers · Test a football situation against graded games. · as of Oct 11, 9:15 PM PDT
- Bench — assemble the model
Snap bricks together into a model, see what each one contributes, and backtest it walk-forward against the same settlement ledger the house is graded on.
1141 building blocks · Build your own angle and test it against graded games. · as of Oct 12, 11:09 PM PDT
- My models — what you own
Every model you have built, with its record, why it passed or failed in words, and whether it is currently clearing breakeven.
435 cards · Follow the walk-forward record of every model. · as of Oct 12, 11:09 PM PDT
- Remixes — start from someone else
Remix a published model, change one thing, and keep the lineage so the two records can be compared honestly.
3 lineage edges · Fork a public model with attribution intact. · as of Oct 12, 11:09 PM PDT
A model's probability is a probability at a price, not a promise. Early samples are marked as early.