Twelve games on the board, three plays clear the model's bar — led by a total in Ottawa with the best expected value of the night.
Twelve games were on the slate today. Three plays cleared the bar. That bar is simple and we don't move it: the model has to like a side at 50% or better, and the price has to return at least 2% expected value at the number you'd actually pay — not some vig-free fantasy line. A play can look fine against a fair number and still be a loser once the book's hold is baked in, and that hold runs roughly 2 to 2.5 points on a typical NHL moneyline. We score against the real price because that's the only price anyone's actually betting.
Three plays made it through today. None of them got padded in to round out a list — this is what cleared, full stop.
Best expected value on the board belongs to the total in Ottawa. The model has the over at 54.6%, against a price that only needs 47.6% to break even. That gap is 14.6% EV, the widest of the day by a comfortable margin — even though this one carries a PASS tier designation from the model's internal grading, worth flagging rather than hiding.
The inputs behind it: home-side injuries sit at -0.3, the away team's 5v5 expected-goals rate against this opponent's defense sits at -0.182, home high-danger rate comes in at +0.057, and the home team's 5v5 xGF against this opponent's xGA sits at +0.049. We're reporting those as the model's raw factor readings — we're not assigning them to a side beyond what's given, and we're not telling you this game will be fast or slow or anything else about how it gets played. That's not information we have.
Two caveats the model itself raises: no goalie is confirmed for either side, and there's no observed GSAx on either netminder. That's a real hole in a total play, and it's why this one's worth sizing with some humility even at a plus-money price.
Kings at Maple Leafs carries the BET tier. Los Angeles on the moneyline at -105, with the model giving the Kings 54.5% against a breakeven of 51.2%. That's 6.4% EV — smaller than the top play, but it's the highest-confidence tier of the three plays that cleared today.
Montréal at Carolina rounds out the board with a LEAN on the over 6.5 at -110. Model number is 55.5%, breakeven sits at 52.4%, leaving 6.0% EV. Lowest of the three plays by expected value, but it still cleared both conditions of the gate.
This is one model's output, run against today's real prices, on a sample that is not large enough to call a demonstrated edge. Treat the tiers — PASS, LEAN, BET — as the model's own internal confidence labels, not a guarantee of anything, and treat all of this as analysis, not financial advice. We're not projecting how any of these games will be played; we don't have line combinations, confirmed goalies, or style notes to hand you, and we're not going to invent them to make the copy read prettier. The numbers are the numbers. Bet accordingly, or don't.