- Win chance
- What Sab thinks the chances are that a team wins, from its own model: team ratings, recent form, rest, travel, injuries and more. Out of 100 games like this one, a 60% team should win about 60.
- Market chance
- What the sportsbooks' odds say the chances are, with their built-in cut taken out.
- Edge
- Sab's win chance minus the market's. +8% means Sab gives the team 8 points more chance than the books do.
- Expected return (ER, expected value)
- What a bet would make on average per 100 points bet, if Sab's chance is right. +5% means about +5 per 100 bet over many bets; it is not a chance of winning. A single bet still wins or loses.
- Fair odds (no-vig)
- The odds with the sportsbook's cut removed: what the price would be if betting were an even game.
- Vig (the book's cut)
- The sportsbook's built-in margin. It's why both sides of a coin flip are priced at −110 instead of +100, and why a bet with no edge loses a little over time.
- Top pick
- The strongest call, from Sab (game bets) or Sib (props). Game bets: at least 3% expected return and 2 points more chance than the market. Props: at least 5% expected return. Members only. Held back while Sab's win chances aren't scoring better than the market's: then it stays hidden until Sab's picks have beaten the closing line over at least 200 of them.
- Value
- A moneyline pick from Sab's own model alone: at least 3% expected return and 2 points more chance than the market. Not a Top pick: it unlocks once Sab's own model has beaten the market in testing.
- Lean / Watch
- Some value, but not enough for a Top pick: 2–5% expected return on a prop (Lean), 1–3% on a game bet (Watch). Worth a look, not a strong bet.
- Pass / Avoid
- No value at this price. Sab wouldn't bet it.
- Not proven on real lines
- Sib's picks on this kind of prop lost money when tested on real past sportsbook lines (300+ picks), so it makes no pick on it until a later test says otherwise.
- Confidence
- How big the expected return is (Low under 5%, Medium 5–10%, High 10%+), plus how Sab's past picks at that level actually did.
- Projection / Projected score
- The guess at the final number: Sab's for the score of the game, Sib's for a player's points, rebounds and so on.
- Chance (on a prop)
- How likely Sib thinks the side it picks is to hit, e.g. "63% chance" for the under. It blends Sib's projection with the sportsbooks' chance (books know about minutes limits and late news); "Sib alone" is the projection before that blend.
- Stake / Kelly stake
- How much of your bankroll a bet deserves. Sab uses a quarter of the Kelly formula (a standard sizing rule) and never more than 5%.
- PICK
- The team Sab thinks is more likely to win this game (during a game: right now, from the score and the clock). It isn't a bet tip on its own: the bet Sab likes is in the Pick section, by value at the price.
