Putting our money where our mouth is.
Model projections vs. the sharp books and Kalshi, every game day — every play graded and published.
1 games · flags: live 0–1–0 (-1.00 u) · finals and graded flags as of the last build
| Kick (PT) | Matchup | Flags | Final | Model line | Pinnacle line | Δ (home) | Line move | Kalshi spread (home · away) | Home win% | Model score | Model total | Pinnacle total | Δ total | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Monday, September 21 · Week 2 · 1 game | ||||||||||||||
| Mon 9/21 5:15 PM | New York Giants @ Los Angeles Rams | modelA New York Giants +7.0 ✗ | 6–28 | LAR -2.7 | LAR -7.0 | -4.3 | flat | LAR -6.5 @51¢ · NYG +7.5 @57¢ | 58% | 22–25 | 47 | 48 | -0.2 | |
Lines name the favorite (e.g. SEA -3.5 means Seattle favored by 3.5). Δ (home) is how many points more the model likes the home side than the market does; negative means it likes the road side more. Line move is our first captured Pinnacle line → the current one, and whether the market has drifted toward the model's side. Kalshi spread shows the closest ladder rung and ask price for each side to cover. Flags are logged plays with the line and price at the time they were flagged: model = our projection disagrees with the market line; price = a Kalshi rung priced better than the market line implies (its rung is deliberately off the line, so only its ask price moves). A = best play, B = lean (measured, not bet). "now" shows the current market line when it has moved since the flag — the edge a follower gets is the difference.