Corsi events per goal

[Cunneen]

I thought it would be interesting to look at how many corsi events it takes for a team to score a goal.
This could be in place of shooting percentage. Rather this stat would be like Total shooting percentage, since it takes into account any time a team misses or blocks a shot. I'll post when I have the results, but let me know if you have any ideas what this data could be used for or what it could be compared to

Click to read more...

 

[Micklebot]

I thought it would be interesting to look at how many corsi events it takes for a team to score a goal.
This could be in place of shooting percentage. Rather this stat would be like Total shooting percentage, since it takes into account any time a team misses or blocks a shot. I'll post when I have the results, but let me know if you have any ideas what this data could be used for or what it could be compared to

I seem to recall it being around 24 corsi events per goal over the last few years, I could be wrong but it's easy enough to compile the data.

In the end, it will get thrown around to try to predict the effect of zone starts, and FO wins and other stats might have on production (ie if a OZ Faceoff win is worth 3 corsi events, and a neutral Zone FO win is worth 1.4 corsi events we should adjust player x's points to y to standardize across the league).

My own opinion is adjustments like the above hypothetical scenario never end up really working out as there are too many variables.

 

[Cunneen]

I seem to recall it being around 24 corsi events per goal over the last few years, I could be wrong but it's easy enough to compile the data.

In the end, it will get thrown around to try to predict the effect of zone starts, and FO wins and other stats might have on production (ie if a OZ Faceoff win is worth 3 corsi events, and a neutral Zone FO win is worth 1.4 corsi events we should adjust player x's points to y to standardize across the league).

My own opinion is adjustments like the above hypothetical scenario never end up really working out as there are too many variables.

I wanted to look at this at first on a team by team basis and then see if Corsi events per goal scored can correlate at all to offensive production for a team. I also would be interested to see if Corsi events per goal is stat that can be produced year after year by good teams, or if it is random (like team shooting percentage).

 

[Cunneen]

so I ran some data for last 3 seasons. I graphed Corsi for/against per goals for/against and Goals for/against per 60 min. The two graphs are attached below. Corsi for/against per goals for/against is similar to shooting percentage, but rather it measures how shots, missed shots, and blocked shots it takes to score a goal (on both offensive and defensive sides).


The R^2 value for both graphs is above .6 No surprise that there is correlation, but just wanted to check that out. Next I want to see if corsi for/against per goals for/against is a repeatable skill or if it is just random (like shooting percentage)

 

[Cunneen]

quick update, my last post was not very useful because I forgot to adjust for save percentage. When adjusting for save percentage (basically finding new goals for and against based off the league average save percentage), the stat becomes Corsi for per goals for adjusted and Corsi against per goals against adjusted. It took me a bit to realize this is just a measure of how many blocked and missed shots a team lets up and takes per actual shot they let up and take.

Now is this valuable, who knows. Hopefully I will figure that out in the next few days (finishing up AP exams in school so can't find time to research this week).

Let me know if you want me to look at anything with these numbers.

Cheers!