The Corsi Rating and Zone Start, Part II

For those of you that wanted to see the Corsi number proven useful, take heart. There is yet another site called Time on Ice that records the shot and faceoff data of every NHL game by harvesting the figures directly from the NHL.com play-by-play reports. This data can be used to figure out what the average “zone start” for each player on the ice.

How is this useful you ask? Assuming we had all of the numbers, our first step would be to discount all of the neutral zone faceoffs. Why, well because they are….neutral. We are only interested in offensive and defensive zone draws.

Our next step would be subtracting the defensive zone faceoffs from the offensive zone faceoffs. Here’s an example of what I mean:

Def. Faceoffs Off. Faceoffs Delta
Player A 500 490 -10
Player B 550 575 25
Player C 475 475 0

The “delta” represents the “zone start” for the player. If the zone start is a positive number then that player was on the ice for more even strength faceoffs in the offensive zone than in the defensive zone. Obviously these players have a better chance of having a higher Corsi number than the players with a negative zone start since they start in the zone most likely to create shots directed to the offensive net.

The next step to acquiring some really useful information would be to chart the zone starts and the Corsi numbers for a group of players. I’m going to borrow this chart from the matchsticksandgasoline.com website which shows where the Flames players chart on the Corsi/Zone Start scale:

As we can see, the players in the upper left quadrant of the graph (Curtis Glencross, Craig Conroy and David Moss for example) are the players that tend to start more often on the “defensive” side of the ice than their teammates but have relatively high Corsi numbers nonetheless. This means that they are able to overcome the disadvantage in ice position and still affect the game in a positive way for their team.

Players that finish toward the bottom right quadrant (Iginla, Cammalleri, and Phaneuf) tend to have a more favorable “zone start” advantage but are not capitalizing on it by directing shots toward the opposition net as frequently as some of their teammates.

Todd Bertuzzi meanwhile had the fifth highest zone start number but the second lowest Corsi number on this chart. This leads me to believe that Bertuzzi was not as effective as he should have been given his situation.

So what can this advanced stat/ratio mean exactly? Well, this type of comparison can show which of your players are good at turning a vulnerable situation into a favorable one. It also shows which players may not be capitalizing as often as they could.

The question is what other factors can contribute to a positive or a negative showing on this chart. First off, notice that three of the lower rated players (Phaneuf, Iginla and Cammalleri) on this chart were also three of the most dangerous players on the ice at any time for the Flames. Subsequently, it is highly likely that the opposing coach had his best defensive players and best faceoff guy on the ice for most of any Flames’ offensive zone draw when Iginla, Phaneuf and/or Cammalleri were on the ice.

The higher the strength of the opponent would seem to be able to offset some of the advantages given by a favorable zone start. Conversely the weaker that the opponent is could also positively impact the ratings of opposing players as well.

Chances are that opposing coaches rarely put their best faceoff or defensive players on the ice simply because the Flames ice Curtis Glenncross, David Moss and/or Craig Conroy for a Flames offensive zone draw. Therefore it may be more likely that the Flames would win the draw and then have the ability to direct shots at the offensive net,  posting higher Corsi Numbers when they were not facing the best faceoff and/or defensive players from the other team.

That same premise would also apply to the top defensive players and faceoff guys whom the coach calls out to take defensive zone draws against the opposing team’s best offensive players. They are already starting in a defensive position by taking the draw in their own zone and then are working primarily to keep the opposing team’s best players from putting the puck in their net. If they can switch positions and get the puck into their offensive zone and create some offensive then that is usually considered “gravy.”

While the addition of the zone start factor enhances the value of the Corsi number, it still falls well short of what an advanced stat should be. The strength or weakness of both the player and their opponents has a huge impact on what this stat can show and therefore diminishes the value of this stat.

However, it is still important not to completely discount this stat as another useful piece of information that can be utilized by NHL GM’s. As was written in a previous “School of Stats” post, all information should be welcomed when trying to properly evaluate a player. Combined with other advanced stats, the Corsi/Zone Start ratings can serve to assist in player evaluation.

That’s it for today, folks. Join me next time for another exciting piece breaking down another advanced statistic or comparing players using an advanced statistic. Please let me know your thoughts on the Corsi/Zone Start ratings specifically or advanced statistics in general.

facebooktwittergoogle_plusredditpinterestlinkedinmail

Comments are closed.