Available in USContent
Ice hockey generates more trackable data per minute than almost any other team sport. Every shot attempt, every zone entry, every faceoff, every shift length is recorded and catalogued. The challenge for bettors is not finding data — it is filtering the useful from the noise. I have spent years building models from hockey statistics, and the single most valuable lesson I have learned is this: most of the numbers people look at first are the ones that matter least.
Ice hockey betting statistics fall into two categories: traditional metrics that tell you what happened, and advanced analytics that tell you why it happened and whether it is likely to continue. Both have a role in a bettor’s toolkit, but the advanced numbers are where the real edge lives, because most casual bettors never look beyond goals and assists.
Traditional Statistics Every Hockey Bettor Should Know
The NHL regular season drew a record 23.16 million fans in 2025-26, and every one of those games generates a box score stuffed with traditional statistics. Here is what each number actually tells a bettor.
Goals and assists are the most visible stats but the least predictive in isolation. A player who scored four goals last week might go scoreless for the next three games because goal-scoring is inherently streaky. Season-long goal totals are more stable, but for game-by-game betting, raw scoring numbers are a lagging indicator. I use them primarily for player prop research — if a forward averages 0.35 goals per game over a season, that baseline informs my assessment of anytime goalscorer props.
Plus/minus (+/-) measures whether a player is on the ice for more goals scored or allowed at even strength. It was once considered the definitive measure of a player’s value, and it is now widely recognised as deeply flawed. A player on a dominant team will have a strong plus/minus regardless of his individual contributions. A strong player on a weak team will have a negative number despite playing well. I mention it only because you will see it in box scores — and to warn you not to overweight it.
Power play percentage (PP%) and penalty kill percentage (PK%) are directly useful for betting. A team with a 25% power play converts one in four advantages; a team at 15% converts one in six or seven. That difference matters enormously for totals betting (more powerplay goals push the total over) and for specific prop markets (powerplay goal scored: yes/no). PK% is the defensive mirror, telling you how vulnerable a team is when shorthanded.
Shots on goal (SOG) is a volume metric that feeds into more sophisticated analysis. A team averaging 33 shots per game creates more scoring opportunities than one averaging 26, all else being equal. But “all else” is rarely equal — shot quality varies wildly, and thirty low-danger perimeter shots are worth less than twenty shots from the slot. SOG is a starting point, not a conclusion.
Advanced Analytics: xG, Corsi and Fenwick
NHL media rights — including the ESPN and TNT deal valued at approximately $4.5 billion over seven years — have fuelled investment in analytical infrastructure across the league. The result is a proliferation of advanced metrics that bettors can access freely. Here are the three that I use most.
Expected goals (xG) is the single most important advanced metric for hockey betting. It assigns a probability of scoring to every shot based on location, angle, shot type (wrist, slap, deflection), whether it was a rebound, and the game state (even strength, power play, shorthanded). A team generating 3.2 xG per game but scoring only 2.5 goals is likely to regress upward — their process is better than their results suggest. Conversely, a team scoring 3.5 goals against an xG of 2.8 is riding unsustainable luck. I compare xG to actual goals over rolling ten-game windows to identify teams whose results are about to shift.
Corsi measures all shot attempts — shots on goal, missed shots, and blocked shots — as a proxy for possession and offensive pressure. Corsi for percentage (CF%) above 50% means a team is generating more shot attempts than it allows, which correlates with winning over time. A team with 55% CF% is dominating puck possession; a team at 45% is getting outplayed regardless of the scoreboard. For betting, a team with strong Corsi numbers but poor results is a value candidate — the underlying performance suggests improvement is coming. The hockey betting strategy guide integrates Corsi and other analytics into a full strategic framework.
Fenwick is Corsi minus blocked shots. The rationale is that blocked shots are a skill — some teams block shots deliberately as a defensive strategy — and removing them gives a cleaner picture of true offensive generation. In practice, Corsi and Fenwick tell similar stories for most teams. I use Fenwick primarily when evaluating teams known for heavy shot-blocking (New Jersey under certain coaching regimes, for example) where Corsi might overstate the opponent’s offensive output.
All three metrics can be broken down by game state: even strength, power play, shorthanded, score-adjusted (accounting for teams that play differently with a lead versus trailing). Score-adjusted metrics are the most valuable for betting because they strip out the behavioural bias of teams protecting leads or chasing deficits, giving you a cleaner read on true ability.
Free and Paid Data Sources for Hockey Bettors
The good news is that most of the data you need is available for free. The challenge is knowing where to look and how to interpret what you find.
The NHL’s official website publishes comprehensive game logs, player statistics, and team data. For traditional stats, it is the definitive source. However, it lacks advanced analytics beyond basic shot data. For xG, Corsi, Fenwick, and scoring-chance models, you need third-party sites.
Natural Stat Trick is the workhorse of free hockey analytics. It provides shot maps, expected goals data, Corsi and Fenwick breakdowns, line combinations, and goaltender metrics — all filterable by game state, strength, and time period. I check Natural Stat Trick before placing any NHL bet.
Money Puck offers expected goals models, playoff probability charts, and team-level analytics with a clean visual interface. It is particularly useful for futures betting because its projections update daily and track how each team’s chances evolve across the season.
Elite Prospects is the primary source for prospect and roster information across all leagues, including the EIHL, European leagues, and junior hockey. When a team calls up a player from their development league, Elite Prospects tells you who they are and what their numbers look like.
EIHL statistics are harder to find in advanced form. The league’s official site publishes basic box scores and standings, but xG models and shot-attempt data are not publicly available at the same depth as the NHL. This data gap is itself an edge — if you attend EIHL games or follow the league closely, your qualitative knowledge fills a hole that data-focused bettors cannot easily access.
What is expected goals (xG) in ice hockey?
Expected goals (xG) is a statistical model that assigns a probability of scoring to every shot attempt based on factors such as shot location, angle, shot type, whether it was a rebound, and the game state. A team’s xG per game estimates how many goals they should have scored based on the quality of their chances. Bettors use xG to identify teams that are outperforming or underperforming relative to their underlying shot quality, which helps predict future results.
Where can I find free advanced stats for NHL betting?
Natural Stat Trick and Money Puck are the two primary free resources for advanced NHL statistics. Natural Stat Trick provides expected goals, Corsi, Fenwick, and goaltender metrics with filters for game state and time period. Money Puck offers team projections and playoff probability models. The NHL’s official website covers traditional statistics comprehensively.
Are Corsi and Fenwick still useful for betting in 2026?
Yes. While expected goals models have become more popular, Corsi and Fenwick remain useful as quick-reference possession proxies. A team with strong Corsi numbers is generating more shot attempts than it allows, which correlates with winning over time. Corsi is particularly valuable for identifying teams whose underlying process is better or worse than their results suggest, making it a useful tool for spotting value bets.
