HOU@DET

This is a tight, low-margin spot where the model recommends HOU +1.5 (-180) despite a virtually flat value picture: the model win prob is 44.0% vs market implied prob 44.5%, leaving a tiny value gap of -0.5%. The market has been drifting toward the home team (opened -122/104, current -138/115) and the home no-vig implied moved from 52.9% to 55.5%, so you’re fighting some movement. With a strength score of 52% the model has modest conviction — this is a small, edges-focused play, not a blowout projection.

The game opened -122/104 and the current line is -138/115, a move toward the home side; the home no-vig implied moved from 52.9% to 55.5% (toward home), so the market has been drifting in favor of the host team. That movement increased the home-side implied probability by 2.6 percentage points (52.9% → 55.5%), which is material in a tight market.

The most realistic way this pick loses is if the market’s home drift (home no-vig implied moved from 52.9% to 55.5%) is driven by sharp money and the sharp_agreement signal 0.211 (contribution 0.042) proves prescient.

No reported injury impact.

The predicted line HOU 3.5 - DET 5 aligns with season scoring: DET averages 4.1 scored · 4.0 allowed and HOU averages 4.4 scored · 5.0 allowed (RECENT FORM), pointing to a sub-5.5 total-type game dynamic.

The strength score of 52% reflects a shallow edge: model win prob 44.0% vs market implied 44.5% yields a value gap of -0.5%, so the model has only modest conviction despite mixed layer signals.

I expect a fairly low-scoring game with Detroit eking out a lead late; projected final score: HOU 3.5 - DET 5.

Bottom line: play the model pick HOU +1.5 (-180) while shopping around — the current sportsbook line listed in the inputs is +115, so get the best available price; this is a small-edge, situational wager.

Shop the price — the inputs show the market moved from opened -122/104 to current -138/115 and the sportsbook line listed is +115, so look for the friendliest number on HOU +1.5 (-180) or better. Consider small unit plays; avoid over-levering correlated parlays here given the modest strength (52%) and the -0.5% value gap.