PropBetting articles › Week 5
Week 5 player props by game: fair odds vs the market
AI-assisted. Written automatically from our model's own numbers with fixed templates; nothing is invented, and no language model adds facts. How this works.
Below, game by game, are the Week 5 props with the largest estimated edge over the posted price. The numbers are the same ones on our props board. Fair probability is our simulation blended with the market (fit walk-forward on past seasons); no-vig is the market's own probability with the bookmaker's margin removed; EV is the expected return per $1 at the posted price if our fair probability is right.
How we label sides: in our out-of-sample backtest of this exact pick rule (fit on 2023-24, tested on 2025), the top unders by edge beat the price; the top overs did not show a significant edge. So every over below is labelled unproven. A backtest is not a live record: the live, graded record is on the Track Record page and in the table further down.
21+ only. This is model output, not betting advice, and no outcome is guaranteed. Bet only what you can afford to lose. Gambling problem? Call 1-800-GAMBLER (1-800-426-2537) or visit ncpgambling.org.
TB at DAL
Top of this game: Ted Hurst III under 1.5 receptions at +154 (FanDuel): our fair probability 40.8% vs 36.8% no-vig, estimated EV +3.8%. Our simulated median is 1.0. (game page)
| Player | Prop | Pick | Price | Fair odds | No-vig | Est. EV | Side label | Live record (stat, side) |
|---|---|---|---|---|---|---|---|---|
| Ted Hurst III | Receptions | Under 1.5 | +154 FanDuel | +145 | 36.8% | +3.8% | Under: backtest-validated | 49-35 (+13.19 units) |
| Jalon Daniels | Rushing yards | Under 43.5 | −114 FanDuel | −124 | 50.0% | +3.8% | Under: backtest-validated ⚑ large model-market gap | 25-11 (+11.07 units) |
| Tyler Goodson | Rushing yards | Under 18.5 | −114 FanDuel | −116 | 50.0% | +0.8% | Under: backtest-validated | 25-11 (+11.07 units) |
| Ryan Flournoy | Receptions | Under 3.5 | −160 FanDuel | −161 | 57.5% | +0.2% | Under: backtest-validated | 49-35 (+13.19 units) |
SF at SEA
Largest edge: Christian McCaffrey under 4.5 receptions at −106 (FanDuel): our fair probability 51.9% vs 48.1% no-vig, estimated EV +0.8%. Our simulated median is 4.0. (game page)
| Player | Prop | Pick | Price | Fair odds | No-vig | Est. EV | Side label | Live record (stat, side) |
|---|---|---|---|---|---|---|---|---|
| Christian McCaffrey | Receptions | Under 4.5 | −106 FanDuel | −108 | 48.1% | +0.8% | Under: backtest-validated | 49-35 (+13.19 units) |
| Sam Darnold | Rushing yards | Under 4.5 | −108 FanDuel | −108 | 48.8% | +0.1% | Under: backtest-validated | 25-11 (+11.07 units) |
Graded so far: Week 4
Official record for the graded Week 4 games, wins and losses alike: 87-62 (+20.09 units) at the posted prices. Unders 86-57 (+23.75 units); overs 1-5 (−3.60 units). That covers 15 games graded so far. Per-game detail is in the game recaps; the official running record is on the Track Record page.
The official record counts each pick's final pre-kickoff version priced on our market-blended fair probability, the way the board showed it. Rows whose final version was still on raw simulation probability are graded separately and are not part of the record: 140-202 (−64.24 units). Rule details are on the methodology page.
Biggest misses (highest estimated EV that lost): Laquon Treadwell under 1.5 catches (actual 5); Kene Nwangwu over 0.5 anytime td (actual 0); Darius Cooper under 1.5 catches (actual 3).
No pick here is certain, and estimated EV assumes our probability is right, which the live record has not yet shown. 21+ only. This is model output, not betting advice, and no outcome is guaranteed. Bet only what you can afford to lose. Gambling problem? Call 1-800-GAMBLER (1-800-426-2537) or visit ncpgambling.org.
Get this in your inbox every week
Free weekly email: start/sit, the props our model rates as unders, and every graded result, wins and losses. 21+.
Model opinion, not advice. For entertainment and informational purposes only. 21+. Gambling problem? Call 1-800-GAMBLER. Methodology · Track record