Unfolding Hockey Analytics

For many fans, terms like Corsi or expected goals feel like something out of a math class, but they simply put numbers on things your eyes already notice during a game. Instead of stopping at goals and assists, these stats show who is driving play and how often a team creates real scoring chances.

This guide walks through the main modern metrics, starting with shot-based measures such as Corsi and Fenwick, then moving on to expected-goals models and a few key goalie- or special-teams indicators. You will also see how these ideas connect to betting, helping you judge moneyline prices and spot value in goal totals. By the end, the core terms should feel less like jargon and more like tools you can use to understand the game a little better.

Understanding the Basics of Hockey Analytics

For many fans who want hockey analytics explained in simple terms, the best starting point is shot volume. These numbers try to answer a basic question in every game: which team is spending more time attacking and forcing the other side to defend. Instead of just looking at the final score, they track how often a team directs the puck toward the net and use that as a stand-in for puck control.

Shot Volume & Possession Metrics

A good first question for any game is which team actually spends more time attacking. Two of the most common tools to judge this are Corsi and Fenwick, which both use shot volume to estimate who is driving play.

• Corsi: This measures shot attempts. It counts shots on goal, missed shots, and blocked shots, then compares attempts for and against. When a club posts around 60% of the attempts in a game, it is a strong sign that it carried most of the play. Over a full season, teams that consistently win the shot-attempt battle usually win more games.
• Fenwick: This works almost the same way but ignores blocked shots and tracks only unblocked attempts. This can be useful because shot-blocking teams can inflate Corsi against without giving up as many shots that reach the net. If a team is above 50% in both Corsi and Fenwick, it usually means it is spending more time on attack and generating more looks than it allows.

These metrics matter because they show who controlled play, and teams that consistently win the shot-share battle usually climb the standings. That is why coaches and bettors use Corsi and Fenwick to judge how strong a team really is.

Shot Quality Metrics (Expected Goals)

Shot totals tell you who is shooting more, but you also want to know which team is getting the dangerous chances. Expected goals take every shot and ask a simple question based on where and how it was taken. How often does something like that go in over thousands of similar plays?

• Expected Goals (xG): This assigns each shot a score between 0 and 1 based on its likelihood of going in. A shot from right in front of the net or on a breakaway gets a high xG, while a low-danger shot from the outside carries very little. Add up a team’s shot values, and you get a picture of how much real scoring threat it created in a game. Exact xG values vary by model, but the idea stays the same.
• xGF and xGA: When you sum the xG for all shots your team takes, you get expected goals for (xGF). Do the same for the chances you allow and the expected goals against (xGA). A team can be outshot but still lead in xGF if its chances came from better areas.
• xGF percentage (xGF%): This is the share of total expected goals that belongs to one team. If you create 2.0 xGF and allow 1.5 xGA, you sit a little above 50%. Over time, teams that stay above that line usually drive more quality chances and post stronger results.

If you want hockey analytics explained in a way that helps you read games, xG is one of the best tools to focus on because it ties shot location and chance quality to how often teams should score over the long run.

PDO (The “Luck” Metric)

Hockey is fast and chaotic, so results can swing on bad bounces or a hot goalie. PDO is a simple way to check how much of a team’s success might be driven by that kind of good or bad fortune. You calculate it by adding a team’s shooting percentage to its save percentage at 5-on-5. On the standard 100-point scale, the league tends to hover near 100 over time.

Over time, most teams drift back toward that 100 mark because league-average shooting and save rates are fairly stable. A very high PDO often hints at a record boosted by heaters, and a very low one points to a slump that may correct.

Goaltending and Special Teams

Goaltending and special teams often decide whether a team survives tight games or lets points slip away, so they get their own attention in analytics. Even when five-on-five play is pretty even, a hot goalie or a sharp power play can make a big difference over a full season.

In the following sections, we examine the key metrics that track the number of goals a goalie saves above expectation and the efficiency of teams in attacking with the man advantage or killing penalties.

Goaltending Metrics – Goals Saved Above Expected (GSAx)

For goalies, one of the clearest advanced statistics is Goals Saved Above Expected (GSAx). It shows how many more goals a goalie stopped compared to what an average goalie would have stopped.

What GSAx Measures

GSAx compares the actual goals a goalie let in to the goals they were expected to let in based on how hard the shots were. It uses expected goals models that look at where each shot came from, how close it was to the net, and what type of scoring chance it was. So a weak shot from far away is counted very differently from a dangerous one-timer right in front of the net.

How GSAx Is Calculated

To get the GSAx, you add up the expected goals against from all shots faced, then subtract the real goals allowed. When that total is positive, the goalie has saved more than expected. When it is deeply negative, more pucks went in than the model predicted, which usually signals rough form or shaky defending in front of him.

How to Use GSAx

Over a full season, a strong positive GSAx points to a goalie who regularly bails out his team, while a significant negative number can expose a weak spot in net. In goalie analysis, GSAx is usually the main reference point for judging true performance rather than just raw save percentage.

Special Teams Efficiency – PP%, PK%, and Special Teams Index

Moments with an extra skater or while killing a penalty often decide tight games, and over a full season, they can separate true contenders from teams that hover in the middle of the standings. Analytics treats these situations separately by tracking how often a club converts with the man advantage and how often it survives when shorthanded.

• Power play percentage (PP%): This metric measures how often a team scores on the power play. You calculate it by dividing power play goals by power play opportunities and turning that into a percentage. League average usually sits in the high teens to low 20s, so anything much higher usually signals a unit that can punish mistakes and change a game quickly.
• Penalty kill percentage (PK%): This shows how often a team gets through a penalty without allowing a goal. You divide successful penalty kills by total penalties taken. Numbers around 80% are typical, and teams well above that mark are really good at frustrating opponents and staying in games even when they take penalties.

Many analysts also look at the Special Teams Index, which is simply PP% plus PK%. A combined number near one hundred is roughly average. When that total lands well above one hundred, it hints that a team is gaining an edge in special teams over the long run, which can quietly help them win more games even if their regular five-on-five play looks average.

Applying Analytics to Betting

Modern bettors lean on analytics to see past simple win-loss records and try to spot teams that are undervalued. Hockey will always have bounces and upsets, so these numbers never guarantee a winning ticket, but they can nudge you toward smarter decisions over time. The basic idea is to compare how a team is actually playing to how the market seems to rate them.

• Moneyline Bets (Picking Winners): Underlying numbers can reveal hidden strengths or weaknesses. A team on a losing streak might still have strong Corsi and xGF% while PDO sits well below average, suggesting they are controlling play but running cold on finishing or goaltending. That type of team could be a good underdog bet. On the other side, a club winning lots of games but with poor possession numbers and a really high PDO might just be getting lucky with hot shooting and great goaltending, which usually doesn't last.
• Puck Line (Goal Spread): The puck line pays attention to margin, so you want to know who actually tilts the ice. Teams that regularly outshoot opponents by big gaps and create much more xG, especially against weak defences or goalies with poor GSAx, are better bets to win by two or more goals (covering -1.5). When both teams play slow, defensive hockey and most games are decided by one goal, betting on the underdog +1.5 can be smarter.
• Over/Under (Totals): For betting on total goals scored, look at how many quality scoring chances each team creates and allows, then also consider goaltending and special teams. Matchups between two high-xGF teams with shaky goalies and dangerous power plays often lean toward higher-scoring games. Games between stingy defences with strong xGA numbers and goalies in good form are better under candidates. Over a long stretch, using analytics this way helps you filter out noisy short-term streaks and focus on which styles and matchups truly support your bet.

Situational Factors to Consider

Numbers tell you a lot about a team, but the situation around each game can bend those numbers in subtle ways. When you look at analytics or place a bet, it helps to add context so you don't treat every game the same way.

• Home vs. Road Splits: Teams often play better at home thanks to the last line change, crowd energy, and a familiar routine. That bump shows up in win rates and in possession numbers. Rink crews also record events slightly differently in each building, which can nudge shot counts up or down. Venue-adjusted stats try to fix these differences so you can compare teams more fairly.
• Back-to-Back Games (Rest vs. Fatigue): A team playing for the second night in a row, especially after travel, tends to skate a little slower and make more mistakes. Coaches may also turn to a backup goalie in one of those games. That mix of fatigue and lineup changes can drag down Corsi, xGF, and defensive sharpness, which is helpful to keep in mind before you trust season averages.
• Score and Venue Adjustment (SVA): Teams that are ahead often sit back and allow more harmless shots, while teams that trail push forward and inflate their attempts. Score and venue-adjusted metrics re-weight those events, giving more importance to shifts in close game states. When you compare clubs using SVA numbers, you get a view that is closer to their true level rather than the style they adopt while protecting or chasing a lead.