The road team +1.5 for every game tonight: our model's chance it covers, with the graded plays marked. A play is any road +1.5 at 65% or higher, settled against the final score the next morning.
Left: the graded record on the performance page. Right: what taking the road +1.5 in every game did, 2,634 of 3,936 regular-season games, 2023-24 to 2025-26.
Every game's road +1.5 probability, with the graded plays marked, updated three times a day. See plans, or read how the pick works below.
See plans →The model estimates one thing: the chance the home team wins by two goals or more. Everything else — a road win, or a home win by a single goal, overtime and shootouts included — is a road +1.5 cover, so the number on the board is the road team's chance of covering +1.5. It is an XGBoost model on team form and game context. It never sees a betting line.
The graded rule is fixed for the season: every road +1.5 at 65% or higher is a play, every one is graded, and nothing below it counts. Read the percentage as a ranking of tonight's games — a 75% is a stronger road +1.5 than a 66% — more than as an exact probability.
The road +1.5 covers most NHL games on its own: 66.9% of them over the last three regular seasons. So the graded record should be read against that, not against a coin flip. A season in the 70s is the model picking the stronger road +1.5 spots; a season near 67% is the always-road base rate.
It is a hit rate, not a return. Because the +1.5 covers so often, books price it well past even money, usually around -200 or longer, and at -200 you need to win 66.7% just to break even. The model is blind to the price, so check the number against what your book is asking.
The pick is always the road team. When the road team is the underdog, its +1.5 is the posted puck line. When the road team is the favorite, the book posts it at -1.5, and the road +1.5 is an alternate line, often priced -300 or longer. In those games the number on the board is not a call that the home underdog keeps it close; it is the road favorite's chance of winning or losing by one.
How often each side of the puck line covers, and why so many NHL games end two or three goals apart, is in NHL puck line explained.
The favorite is -1.5 goals and the underdog +1.5. The +1.5 side wins the bet if its team wins the game or loses by exactly one goal. Overtime and shootout games always end one goal apart, so the +1.5 side always covers them. The -1.5 side needs a win by two or more.
Our model estimates the chance the home team wins by two or more goals. Everything else, a road win or a one-goal home win, is a road +1.5 cover. When that chance of covering is 65% or higher, the road +1.5 is a graded play, settled against the final score.
In 66.9% of games over the last three regular seasons: 2,634 of 3,936 from 2023-24 to 2025-26. That is the baseline our graded record should be read against, not 50%. A road +1.5 play that covers two times in three is doing what the market already expects.
Not by itself. A +1.5 that covers two times in three is usually priced around -200 or longer, and at -200 you need to win 66.7% just to break even. Our model never sees the price, so the record is a hit rate. Compare it with the price your book is asking.
The pick is always the road team. When the road team is the favorite, its +1.5 is an alternate line, often priced -300 or longer, rather than the posted underdog line. Check which side your book lists at +1.5 before you read the number as a close-game call.
Last week's graded record — wins and losses — and the plays for the week ahead. One email, no daily noise, unsubscribe in a click.
Moneyline, puck line and totals for every game, plus shots, goals, points, assists and saves. MLB, NFL and NBA included.
See plans