An Everyday Guide For Advanced Metrics In Hockey

Published on August 10th, 2014

advancedstats

By: Casey Johnston (@DarthHockey)

The prominence of hockey analytics is growing each season as more people begin to see the usefulness of “fancy stats.”  It’s no different here at The PensNation.  As advanced stats inform our writing more and more, we thought it would be good to compile of list of the more commonly used metrics and give a little insight as to what exactly each means.  This list is by no means exhaustive and it will grow just like the use of analytics is growing.  You can find more stats in the links below.  Some of them even have self-explanatory names!

As always, thanks for reading!

Corsi:

This is the total number of shot attempts by a team or player.  It is almost always shown at 5-on-5.  Unlike the NHL’s traditional shot stat, this includes missed shots and blocked shots as well as shots on goal and goals.  Former Buffalo Sabres goaltending coach Jim Corsi created it as a way to measure how much work a goalie did in a game.  Today, it’s used as a way to measure puck possession.  It can either be presented as a percentage of shots taken (like on Extraskater.com) or a differential number (like on Behindthenet.ca).

Fenwick:

This stat works much like Corsi except it does not include blocked shots.  In a bigger sample size the Fenwick number is considered the more accurate while Corsi is better in smaller sample sizes.

Close-Corsi and Fenwick-Close:

is a team’s Corsi and Fenwick number while the score is tied or a team is leading or trailing by one goal in the first or second period.  The theory is that if a team is leading by more than one goal they may go into a defensive shell to protect the lead.  If a team is trailing by more than one goal they may take more shot attempts.  Close numbers take out that effect.

Relative Stats

Relative stats show how a player performs on the ice versus how his team performs when he’s not on the ice.  In 2013-14 Sidney Crosby had a Corsi Relative of +6.9%, meaning the Penguins Corsi number went up 6.9% when he was on the ice.  Marcel Goc had a Corsi Relative of -11.3%, meaning the Penguins Corsi number was 11.3% lower when he was on the ice.

PDO:

This is the sum of on-ice shooting percentage and on-ice save percentage.  It works as a player or a team stat and it regresses heavily to the mean of 1000 over time.  In other words, a team on a winning streak with a PDO well over 1000 can be expected to regress.  This is especially true if the high PDO is driven by a higher than normal shooting percentage.

Quality of Competition (QoC):

There are several ways to look at quality of competition.  One is to look at the Corsi Relative of the opponents faced by a player.  Another way is to look at the percentage of a team’s ice time that an opponent gets.  This is a way to see whether a player is getting “tough minutes.”

Zone Starts:

This shows where a player starts his shifts.  It’s usually shown as Offensive Zone Start %.  The lower the number the more a player started his shifts in the neutral or defensive zone.  Zone Starts and Quality of Competition give an indication of how “tough” a player’s minutes are.


Here are links to some great hockey analytics sites:
War on IceBehind the Net | Hockey Analysis

 

 

0 comments