NHL Best Bets, Predictions & Expert Picks Friday, October 23, 2026

Analyse

This game is a pass in the spread, total, and moneyline markets, so there is no reason to manufacture a side. Chicago is listed at 3 and 6, while Montréal enters at 4 and 3. The model projects a 6.38 total against a market number of 6.5, but the listed edge rounds to zero. The same applies to the spread and moneyline, both of which carry a pass recommendation. The actionable number comes from Oliver Kapanen's shots on goal prop. His projection is 1.81, which sits above the 1.5 line, but price matters more than the raw projection here. BetOnline's no vig market gives the under a fair chance of 46.9 percent, equivalent to fair odds of plus 113. DraftKings is offering plus 120, creating a 3.3 percent expected value on the under. That is the entire case, and it's a clean one: the available price is better than the sharp fair price. Montréal is also without Nick Suzuki, but the supplied data doesn't quantify how that changes Kapanen's role or shot volume. I won't pretend it does. The full game is a pass, while the prop offers the only documented value. Take the number, not the narrative.

À contre-courant du public

The public may look at Kapanen's projection of 1.81 and instinctively prefer the over. That's the wrong way to attack this market. The sharper signal is the price, with plus 120 beating BetOnline's fair under price of plus 113.

Analyse

Tampa Bay owns the better record at 7 and 2, compared with Colorado at 5 and 4, but the model still doesn't identify a wagerable edge in this matchup. Colorado is the home team, and the moneyline model gives the Avalanche a 53.5 percent win probability. That projection isn't enough to support a recommendation at the available price, so the market gets a pass. The puck line tells the same story. Colorado's model spread is minus 0.19, while the listed line is Colorado minus 1.5. That's a meaningful gap between the model's expected margin and the market requirement. Asking Colorado to win by two goals is not supported by the projection provided. Tampa Bay's stronger record makes the Lightning tempting as an underdog, but no moneyline edge is listed, so there isn't a data backed reason to make that bet either. The total is just as tight. The model projects 6.42 goals against a line of 6.5, yet the listed edge is zero and the recommendation is pass. Colorado will be without Nicolas Roy and Zachary L'Heureux, but the supplied information doesn't assign a specific scoring or pace impact to either absence. That leaves us with a competitive game and no approved wager. Sometimes the sharpest move is leaving a close number alone.

À contre-courant du public

Casual bettors may chase Tampa Bay because the Lightning are 7 and 2, or back Colorado because the Avalanche are at home. Neither record based angle has a listed edge. The model sees a near coin flip, not a bargain.