Among NHL prop markets, shots on goal (SOG) is the analyst's favourite — and for good reason. A shot happens whether or not it beats the goalie, so the market strips out the single biggest source of hockey variance: goaltending. What's left is far more predictable — ice time, role and shot rate. That makes SOG props a rare corner of hockey betting where a model has a real, repeatable edge.
This guide covers the main NHL prop markets, how to analyse a shots-on-goal line the way a model does, and how ProbWin prices shots and points props. As always, the analysis is the product: no guarantees, just where the value is and a public record of every result.
What are NHL player props?
The rule
A player prop is an over/under on one skater's stat. Nathan MacKinnon over 3.5 shots on goal: "over" wins at 4+ shots, "under" at 3 or fewer. The markets that reward analysis:
- Shots on goal (SOG) — high-volume, role-driven, the cleanest to model.
- Points (goals + assists) — higher upside, more variance.
- Assists — the noisiest, teammate-dependent.
ProbWin models shots and points — the markets where ice time and shot rate give a real signal.
Why shots on goal is the modeller's market
A goal prop lives and dies on the goalie and puck luck. A shot prop doesn't care if the puck goes in — only whether the player attempted it on net. That removes goaltending variance and leaves a system driven by minutes and shot volume, both of which are stable and estimable. It's the closest hockey gets to a "clean" prop.
Rejoignez ProbWin gratuitement
Accédez à nos prédictions IA pour MLB, NHL et NFL. Inscription en 30 secondes.
The inputs that move a shots prop
1. Ice time (TOI)
Shots scale almost linearly with time on ice. A top-line forward playing 20 minutes shoots far more than a checking-line grinder at 12. Projected TOI — including whether the player is on the top power-play unit — is input number one.
2. Shot rate (shots/60)
How many shots a player generates per 60 minutes is his underlying volume. Some players are shot machines regardless of role; others pass first. Shots/60 × projected minutes is the backbone of the estimate.
3. Power-play role
Power-play time inflates shot volume dramatically for the players who man the top unit. A shooter on PP1 has a structurally higher line than his even-strength rate suggests.
4. Opponent and game context
Teams that concede a high shot volume (and trailing game scripts, which pump shots from the chasing team) lift a shooter's line. A likely trailing team means more desperate shot attempts late.
How ProbWin models NHL props
Our NHL props model projects a shot distribution (and a points distribution) for each skater from projected ice time, shots/60, power-play role and the opponent's shots-against profile — then converts it into over/under probabilities and compares them to the market. It publishes picks on shots and points specifically, the markets our data supports.
The edge is calibration and discipline, not the concept: when the model says 57% over on a shots line, the over should land near 57 times in 100 over a large sample, and we only surface a pick when the best available line beats our fair line. Every pick is timestamped and settled publicly — winners and losers alike. We do not publish anytime-goalscorer picks, because a shot model and a goal model are different animals and we only sell what we can estimate.
Strategies that hold up
Strategy 1: volume shooters in trailing scripts
A high shots/60 forward on a team likely to trail is a double tailwind: his own rate plus a game script that pumps shot attempts. The over line often lags this.
Strategy 2: power-play role changes
When an injury or line shuffle promotes a shooter to PP1, his shot line jumps before the market fully adjusts — the same "usage vacancy" logic as basketball, applied to special teams.
Strategy 3: fade thin-minutes overs
A bottom-six skater with a tempting over line is usually a trap: not enough ice time to clear it consistently. Minutes beat reputation.
The core discipline
A prop is a bet only when the best line beats the model's fair line, per the value betting principle. Shots are lower-variance than goals, but the value filter still governs every play.
Case study: a shots prop, step by step
The situation
A top-line winger, line at 2.5 shots on goal. A teammate's injury moves him onto the top power-play unit for tonight. Our model lifts his projected TOI and PP share, producing:
| Shots on goal | Model probability |
|---|---|
| Over 2.5 | 60% |
| Under 2.5 | 40% |
Fair line vs market
| Selection | Best market odds | Implied prob | Our prob | Value? |
|---|---|---|---|---|
| Over 2.5 | 1.85 | 54.1% | 60% | Yes (+5.9 pts) |
| Under 2.5 | 1.95 | 51.3% | 40% | No |
Takeaway
The value is on the over — created by the power-play promotion, a role change that raises shot volume before the opening line catches up. As with NBA usage vacancies, the recurring edge is a role shift the market prices late, caught by a model that re-projects minutes and role instantly.
Where NHL prop edges live
| Situation | Best market | Why |
|---|---|---|
| Volume shooter, likely trailing | Shots over | High shots/60 + chase-script shots |
| Promotion to PP1 | Shots / points over | Special-teams time lifts volume |
| Bottom-six thin minutes | Shots under / fade over | Not enough ice time |
| Elite line vs weak defence | Points over | More high-danger chances |
Pitfalls to avoid
1. Confusing shots with goals
A shots prop doesn't need the puck to go in. Don't let goalscoring narratives (or the goalie matchup) sway a shots bet — they're a different market.
2. Ignoring line and PP changes
Hockey rotations shift game to game. A player's line from last week may be based on a role he no longer has. Check tonight's projected lines.
3. Trusting season averages over role
A skater bumped up (or down) the lineup is not his season shot rate. Current role and TOI beat the average.
4. Overstaking points props
Points carry goal variance; a "great" points over can miss on a quiet night. Size points props smaller than shots props.
Frequently asked questions
Which NHL props does ProbWin cover?
Shots on goal and points — the markets our model projects from ice time and shot rate. We do not publish anytime-goalscorer picks, as that requires a different model we don't sell.
Does the goalie matter for a shots prop?
No — a shot on goal counts whether it's saved or scored. The goalie is irrelevant to SOG; it matters for goals and totals, not shots.
Do overtime and shootouts count?
Shots in overtime count toward SOG props; shootout attempts generally do not. Check your book's rule, especially for tight lines.
Are NHL props profitable long term?
Only with a calibrated projection and a strict value filter. Shots are lower-variance than goals, which helps — but the discipline of best-line-vs-fair-line still decides profitability.
When do NHL prop predictions resume?
Our NHL models run through the season (October to June). In the off-season the projections pause; the method and this guide stay live year-round.
Conclusion
Shots on goal is the NHL prop market built for modelling: it removes goaltending variance and leaves a system driven by ice time and shot rate — both stable, both estimable. The recurring edge is a role change (a PP1 promotion, a trailing script) the market prices late, and a model that re-projects instantly gets there first.
The rule holds across every prop: a line is only as good as the projection behind it. That's where our calibrated shots and points models and our multi-book comparison do the work. See today's NHL predictions and our value betting method.
