1. Sports
Send to a Friend via Email

Your suggestion is on its way!

An email with a link to:


was emailed to:

Thanks for sharing About.com with others!

You can opt-out at any time. Please refer to our privacy policy for contact information.

Corsi, PDO and Fenwick: 3 hockey stats you need to know

So what is Corsi?



Patrick Kane and Blackhawks dominate some of the most important statistics in the NHL.

Getty Images Sport

Hockey has a lot of numbers, and some are much better than others when it comes to evaluating the play on the ice.

We already discussed four statistics that you should probably forget about, so now it's time to move on to the numbers you should know, all of which could be considered hockey's version "advanced" statistics. But don't let that name fool you, there is a reason the quotations are there -- hockey's advanced statistics really aren't all that advanced.

They're not only quite simple and easy to track, but if you can grasp the concept behind plus/minus (and every hockey fan can) then you've already nailed the basic concept behind Corsi and Fenwick. 


What it is: It's just like plus/minus, only instead of counting goals for and against it counts total shot attempts for and against. Goals, saves, shots that miss the net, and shots that are blocked. Like most of hockey's newer stats it is named after the person that brought it to prominence, and in this case it was Buffalo Sabres goalie coach Jim Corsi who was looking for a way to measure the workload his goalies had to face during a game. The thinking being that every shot attempt, whether it reached its intended target or not, required a reaction from the goalie. 

Along with that, it's also a pretty good measure of puck possession and how much time a team or player is spending in each end of the ice. A player or team with a high Corsi is going to be spending more time in the offensive zone on the attack, while a player with a negative Corsi is going to be trying to defend and is constantly chasing the puck. 

The NHL no longer publishes zone times in its box scores, so this is probably the best measure we have for determining puck possession. 

Why it matters: Possession and the ability to have the puck more than your opponent is a vital part of winning championships in hockey. It also has more predictive value and is more repeatable than plus/minus which is heavily impacted by goaltending and luck. Teams and players have an impact on the number of shots they generate, but they don't always control how many of those shots or which ones go in the net or stay out of the net.

Still, It's not perfect. When it comes to individual players, a players usage and his role needs to be taken into account. A player that is put into defensive roles (starting most of his shifts in the defensive zone and against better competition) is probably going to see his Corsi numbers take a hit, especially when compared to a player that plays softer minutes (more offensive zone starts, going up against weaker competition). 

There is also the fact that some players, no matter how good they are at advancing the play from their own end of the ice into the offensive end of the ice, just don't have the type of scoring ability (the shot or the ability to find the open spaces in the offensive zone) that can consistently turn it into goals.  

If you were to look at the 2013-14 season and say that Brad Marchand is having a better season than Sidney Crosby because he has a better Corsi number you would be wrong. But if you said that Marchand is a good, valuable two-way player that helps his team drive possession and keep the puck out of their own end of the ice, that would be pretty accurate. 


What it is: To understand FenClose we first must understand what Fenwick is. And just what is Fenwck? Well, it's pretty much the same thing as Corsi, only it removes blocked shots from the equation. Why remove blocked shots? Because shot blocking is something that is a skill that players can directly impact.

From there, we can break it down to FenClose, which is the percentage of unblocked shot attempts a team takes in a game when the score is close (within one goal or tied). Look at it this way, if Toronto and Montreal are playing and the two teams combined to take 100 unblocked shot attempts with the score close, and Toronto had 38 of those attempts, Toronto would have a FenClose percentage of 38 percent.

Why do we care if the game is close? Because when teams get ahead or behind by two or more goals they tend to change the way they play (even if it's not always intentional), especially as it gets to be later in the game. A team that has a two-or three-goal lead in the third period is going to play a more passive, careful game than a team that is trailing by the same margin. When the game is close, or even tied, teams are playing more within their system and it's a better reflection of their true talent level.

All of this is referred to as score effects. 

Last week's Carolina-Columbus game, where Carolina took a 3-0 lead to the third period, is a good example of score effects at work. Any lead after two periods greatly increases your chances of winning, but a three-goal lead at that point is one that you can usually put in the win column, and Carolina played that way, completely taking its foot off the gas. Every time the Hurricanes hit the red line with possession of the puck they simply dumped it into the offensive zone and retreated back in an effort to play it safe and not make a mistake. As a result, Columbus, which was in desperation mode and trying to fire shots at the net from every angle, ended up outshooting Carolina 19-0 over the final 20 minutes. If the game is closer neither teams plays that way and the shot numbers don't look anywhere near as lopsided over the final 20 minuets.  

We have FenClose data as far back as the 2007-08 season. Since that time seven of the 12 Stanley Cup Final teams (including four of the six champions) have finished the regular season in the top-5 in FenClose, while only one team (the 2007-08 Penguins, which lost the Final to Detroit) finished lower than 14th (they were 27th and lost to a Red Wings team that was No. 1). 


What it is: Remember when I mentioned luck back in the Corsi section? Well, if there was ever a way to measure whether a team or player has been lucky or unlucky, PDO might be it. And it's not all that hard to figure out as it's simply adding even-strength save percentage and shooting percentage. If nothing else, it's a quick and easy way to look for teams and players that are maybe riding a hot streak and playing over their talent level over a short period of time, or teams that are riding a cold streak and are due to bust out. 

But don't some players and teams shoot at a higher percentage than others, you're probably asking? Of course they do. Phil Kessel is going to score on a higher percentage of his shots than Craig Adams is. Alex Ovechkin will have a higher shooting percentage than Daniel Winnik or any other third-or fourth-liner. But if a player that is an eight or nine percent shooter for his career and suddenly has a season where he shoots at 18 or 20 percent, it's a good bet that guy is going to see his numbers come crashing back down the next season. 

It works the other way as well. A couple of years ago Anaheim's Ryan Getzlaf had a terribly unlucky season offensively. A 12 percent shooter for his career, Getzlaf finished the 2011-12 season by scoring on just 5 percent of his own shots while the Ducks, as a team, scored on just 7 percent of their total shots with him on the ice, leading to one of the worst seasons of his career when it came to goals and assists. It was also the worst season of his career percentage wise with a career-low 977 PDO. It was an outlier in his career in every way, and a season that was crushed by bad luck.

Once his luck changed over the following two seasons he's been back to producing at an All-Star pace. 

On a team level, there is very little difference in team shooting percentage when it comes to the best and worst teams. Over the past six years (combined) the Pittsburgh Penguins have the best 5-on-5 shooting percentage in the league at 8.71 percent (and given the talent they've had on the roster it probably shouldn't be a surprise they're at the top of the list). The worst shooting percentage over that stretch? New Jersey at 7.20. That means over a six-year stretch the difference between the best shooting team and worst shooting was less than two percent. For every 1,000 shots on goal, you're probably looking at an additional 15 goals between the best and worst team. It could definitely be the difference between a couple of wins over the course of a season, but it's not a huge gap. 

That is why shot volume (Corsi, FenClose) matters so much. 

Where can you find these stats? Any number of resources around the Internet, including Behind The Net, Extra Skater, and Hockey Analysis.

©2014 About.com. All rights reserved.