Right now the NHL is in the midst of an "advanced stats" revolution, highlighted by the hiring of several prominent "advanced stats" figures by NHL teams over the past two years. Don't want to feel left out when your friends start talking about Corsi, WOWY's, or HERO charts? Here's your one-stop shop to breakdown what each stat measures and how it should be used appropriately.
What does it measure?: Corsi, or Shot Attempts (SAT), measures exactly what it says it does - the number of shot attempts taken either by an individual team or an individual player. In a math formula it is "Corsi = shots on goal + missed shots + blocked shots". You'll often see Corsi reported as a Corsi For% for a team which is the number of shot attempts for a team divided by the total number of shot attempts that occurred. A percentage above 50% indicates that a team is taking more shot attempts than their opponents.
How is it used? In the analytics world, think of Corsi as a surrogate for possession. Since we can't measure the exact number of seconds a team has the puck easily, we use Corsi as a pretty good proxy. The more shot attempts a team takes, the more they must have the puck. We use Corsi to say that a particular team or player is "better" at puck possession than another team or player.
Example sentence you might see: Pavel Datsyuk and Tomas Tatar are two of Detroit's most dominant puck possession players, having 5v5 CF%'s of 59.63% and 59.01%, respectively. What this sentence shows is that when Datsyuk is on the ice, 59.63% of the 5v5 shot attempts that occur between both teams are for the Red Wings.
What does it measure? Fenwick as a math formula measures "Fenwick = shots on goal + missed shots". Fenwick leaves out blocked shots as it's thought that blocking shots isn't exactly a true "skill" and therefore analytics guys want to remove that element of "luck".
How is it used? Fenwick isn't really widely used much more in the analytics community as most people will opt to use Corsi.
Example sentence you might see: Brendan Smith led all Red Wings defensemen in Fenwick For% at 54.14% despite having more than 37% of his individual shot attempts blocked.
What does it measure? PDO as a math equation measures "PDO = Save% while a player/team is on the ice + Shooting percentage for a player/team when on the ice". For a team, this is just the team's save percentage + team shooting percentage. For an individual player, it's the shooting percentage of the player + the save percentage of the team while that player is on the ice.
How is it used? PDO is essentially used to represent how "lucky" or "unlucky" a player/team has been. A standard PDO is consider to be 100. Anything that's greatly above 100 indicates that the player/team has been lucky or may be on an unsustainable run while the a number below 100 indicates the opposite
Example sentence you might see: Gustav Nyquist's 5v5 PDO of 105.75 in 2013-2014 clearly shows why his overall play regressed a little in the 2014-2015 season as that number was bolstered by his ridiculous 18.75% 5v5 shooting percentage.
What does it measure? Zone Starts are exactly what they sound like - they show how many times a player started the shift in the offensive, defensive, and neutral zones. You'll often seen Zone Starts represented as ZSO% which is showing taking all the players offensive starts and dividing it by the total number of offensive and defensive zone starts a player had. This percentage essentially shows if the player is starting more shifts in the offensive zone or in the defensive zone.
How is it used? You'll see ZSO% used to illustrate a player's "difficulty of shifts". If a player has a ZSO% less than 50%, that player starts more shifts in the defensive zone than in the offensive zone and is perceived to have a more difficult shift. Right now, there is some work being done to challenge this notion by suggesting that zone starts only affect the first ten seconds of a player's shift, but for right now, you can use zone starts to give yourself an idea of where a player most often starts their shifts.
Example sentence you might see: Luke Glendening was challenged with massive defensive responsibilities in the 2014-2015 playoffs, posting a ZSO% of just 15.91%.
What does it measure? Rate statistics measure what a player or team is doing per 60 minutes. It's what would happen if that player/team played the entire 60 minutes of a game.
How is it used? These statistics are sometimes used to help account for one player playing a lot more minutes than another player. They are also used to analyze the effectiveness of powerplays as it's tough to do based on raw numbers since powerplays account for only a few minutes of each game. The per/60 minute rate stats are most frequently used with Corsi, Fenwick, and other shot-based metrics but they can technically be used for anything.
Example sentence you might see: The Wings really struggled defensively at 5v5 in the playoffs when Drew Miller was on the ice, allowing a whopping 60.78 CA/60.
What does it measure? Relative statistics measure what a player is doing relative to the other players on his team. Essentially, how much does a player do a particular thing better or worse than the rest of the players on his team.
How is it used? Relative statistics are most often used to compare players on the same team. If I were to look at the Corsi Rel% for two players on the same team, I could suggest which player is performing better on the same team. These stats are useful because we don't have to worry as much about uncontrolled variables such as team systems.
Example sentence you might see: In the playoffs, Luke Glendening took the lion's share of defensive zone starts for the Red Wings with a whopping 46.50% ZSO%Rel.
What does it measure? WOWY stands for "With Or Without You". WOWY statistics show how well or poorly a player plays with each individual player.
How is it used? WOWY is used most often to show which players play extremely well together or which player combinations should be avoided.
Example sentence you might see: Based on Datsyuk's WOWY chart (shown below), you can see that he elevates the level of play of just about every player that plays with him.
What does it measure? HERO stands for Horizontal Evaluative Rankings Optic. Early, we mentioned what a player's Corsi For% is. Many feel that each player's Corsi statistics and Goal Statistics are affected by the quality of teammates they play with, the quality of competition each player faces, and their zone starts. Well this chart adjusts for all of those variables to give you what each player would look like if their "usage" was adjusted. The charts are the totals for each player since the start of the 2012-2013 season
How is it used? HERO charts are one of the most effective ways to compare players from different teams as the charts account for the differences in their usage. They can also be used to help justify why a player should receive more ice time based on the fact that they are producing like a higher-level player.
Example sentence you might see: While Tomas Jurco's individual raw numbers don't look great, his usage-adjusted puck possession numbers are great and he should receive more minutes.
Recommended Websites To Use: Own The Puck
I hope this helps! Remember that statistics should play a supporting role in your thought process and that pooling together many of these statistics can strengthen your argument. If you have any questions on other statistics you would like explained or would like me to elaborate on anything I've posted here, please post in the comments and I can address these!