TOR@SF

Take San Francisco +1.5 (-175) at the listed sportsbook price: the model projects Toronto 4.3 — San Francisco 2.8, with a model win probability of 47.0% versus a market implied 47.4%, leaving a small value gap of -0.4%. The core EV angle is that the game projects low-scoring (combined 7.1 runs) and the +1.5 cushion neutralizes the slight downside while preserving upside on a tight game.

The market opened -102/-116 and is now 102/-122; home no-vig implied moved from 48.5% to 47.4% (toward away), indicating the line has shifted against the home side since open.

The most realistic way this loses is sharp money and model signals run against the pick — sharp_agreement signal -0.086 with contribution -0.017 suggests professionals favor the other side.

No reported injury impact.

Both teams' season scoring lines are similar (SF Avg 4.1 scored · 4.8 allowed; TOR Avg 4.0 scored · 4.5 allowed), which supports the model's low total prediction TOR 4.3 - SF 2.8 and a combined 7.1 runs.

The strength score of 52% reflects a modest tilt where the model's internal signals and situational layers slightly favor the pick, even though the value gap is -0.4% (model 47.0% vs market 47.4%).

This plays out as a low-scoring game where Toronto edges more runs but not enough to beat a +1.5 cushion — final expected line: Toronto 4.3, San Francisco 2.8.

Play San Francisco +1.5 (-175) at the listed sportsbook price (+102) as the primary ticket; it's a small, pragmatic wager that protects against a one-run loss while aligning with the model's low-scoring projection.

Shop the price and take SF +1.5 at the best available +102 or better; if juice drops or you find a lower -175 vig on the +1.5, move to secure it. Consider correlated small parlays that pair SF +1.5 with unders if you can get a favorable total, but do not increase ticket sizes given the Value gap of -0.4% — this is a modest, hedged expectation, so keep stakes proportional.