Squawka Signal is our in-house NHL model. It gives every team a probability — to make the playoffs, win its conference and win the Stanley Cup — and a projected points total. We then set those numbers against the live prediction market, so you can see where our model and the money disagree. Here is how it works, in plain terms.
Team power ratings
At the core is a single strength rating for every team, built with the Elo system — the same method chess uses and FiveThirtyEight popularized for hockey. Win and you take rating points off the team you beat; win by more, or beat a stronger side, and you take more. Our ratings are built on the FiveThirtyEight framework and carry more than a century of league history, and they update after every game. Home ice is built in. At the start of each season every team is regressed partway back toward the league average, because last year’s form only tells you so much about this year’s roster. We validated the engine against FiveThirtyEight’s own published ratings and it matches them essentially to the decimal.
Overtime, shootouts and the loser point
Hockey has a wrinkle other sports don’t: a team that loses in overtime or a shootout still takes a point. The model handles that the way the standings do — two points for a win, one for an overtime or shootout loss — so the projected points totals are on the same scale you see in the table. And because a one-goal overtime result is largely a coin-flip, the model doesn’t reward a shootout win the way it rewards a blowout; how you won matters, not just that you won.
Simulating the season
From those ratings we play out the rest of the season ten thousand times. Each simulation runs every remaining game, then builds the real playoff field — the top three in each division plus two wild cards in each conference — and plays the full best-of-seven bracket round by round to a Stanley Cup champion, with home ice going to the better regular-season side. Count how often each team reaches each stage across all ten thousand runs and you have its playoff, conference and Cup probabilities, plus a projected points total. It refreshes as results land, so the numbers move with every night’s games.
Why the favorite isn’t higher
A team rating is a best estimate, not a certainty, so we don’t treat it as exact. Every simulation nudges each team’s strength up or down by a random amount before the games are played, which spreads the outcomes realistically. In hockey that spread is wide: a single hot goalie can win a series, so even the best team in the league sits around one in eight for the Cup rather than running away with it. We tune that uncertainty directly against the market — the dial is set so the overall shape of our title race matches what Kalshi and Polymarket are actually paying, rather than making our favorites look artificially strong.
What we publish
For the season the model gives each team’s playoff, conference and Stanley Cup odds and a projected points total. We set the Cup odds against the live prediction-market price — a de-vigged, volume-weighted blend of Kalshi and Polymarket — and flag the gap: where our probability sits above the market we call it value, below it the market is warmer than us. It is a read, not a tip.
What it does not do
A word on what the model is not. It rates teams on strength and results and simulates the whole season; it does not yet price which goalie starts on a given night, and in hockey that is the single biggest thing a pure team-rating model can miss — the starting goalie swings games and series more than any other player. A goaltending and advanced-metrics layer is in the works to sharpen exactly that. It is not a tip service, and on plenty of teams it will simply agree with the market. Treat it as one informed input, not a guarantee.
More from Squawka NHL Signal: NHL Signal hub · Stanley Cup odds · Power Rankings · Standings · Stat Leaders · AI Predictor.