Recently I’ve been exploring the concepts of shot distance, how they relate to shot quality, and what that can tell us about teams and players.

In my game summaries, I’ve added statistics on the average shot distance of unblocked shots and the ratio of shots from within certain distances at the team and player level. There seems to be a renaissance of thought underway about shot quality, and I think it’s important generally to be able to tell the difference between harmful shot situations and more harmless ones. I started documenting these stats purely as a matter of curiousity, but now that I have I’ve gone down many other avenues in my thinking.

Firstly, I needed a lot of data to test theories. The easiest place to get my hands on a ton of event data was from last year’s Hackathon file supplied by the Oilers. I can only open about 1.05 million rows of that file in Excel before it exhausts my computer’s memory, which is just over 2 and two-thirds seasons of data spanning 2007-08 to 2009-10. I’ll get my hands on more complete data in the future, but this is what I have to work with right now — in any case, having over 219,000 Fenwick events to analyze is probably an ok starting place.

First I wanted to find shooting percentage (and therefore, save percentage as well) by shot distance to see whether some earlier studies I’d read on the topic held. Here’s a graphic:

We can see that there’s an obvious relationship between shot distance and scoring goals. Without getting into the details about the kinds of shots, rebounds, etc, I don’t think it’s a controversial proposition to suggest getting closer to the net to take shots is better than taking shots from further out. With all the effort we put into studying shot differentials, I think this line of study is a natural complement.

I think of each offensive attack (or sortie) as an effort in getting into the best scoring position possible before attempting to get a shot off. Many times, this involves getting closer to the net, like a rugby team getting closer to a try. When you’ve exhausted your options in getting any closer, you shoot. Logic would suggest that bad teams and players do poorer in this objective of getting closer to take shots than good ones.

Getting closer not only aids in a shot on net going in, it also helps your shot attempt to hit the net:

This chart shows what percent of unblocked shot attempts count as a shot on net by shot distance. The closer you are to the net, generally, the higher your odds of hitting the net are.

Now just think about those two engines for a second:

- Closer shots are more likely to hit the net
- Closer shots that hit the net are more likely to be goals

These two forces compound on each other quickly to end in pretty dramatic results:

Every foot of ground you can gain before letting your shot go increases your odds of scoring in a non-linear fashion. 8.7% of shot attempts from 20 feet result in goals. That percentage increases to 11.4% at 15 feet. But if you take a shot at 10 feet, it has a 17.4% chance of going in.

I’ve been playing around with a metric that tracks the rate at which a team or player creates or allows Fenwick events within 25 feet. I just arbitrarily picked 25 feet as that seemed like a pretty intrinsic point, as a goalie myself and a spectator, at which shots really start to become dangerous. My first graph above shows that the point at which shots start to go in at below league average is about 27 feet. Anything nearer than that and shots go in at above league average, or looked at in a certain sense, it’s the point where shots become dangerous.

I would agree that the 25 foot threshold is too arbitrary to use very broadly, but in my mind it answers the specific question of how many “close” shots a player creates or allows. Earlier in the season I’d wonder if Justin Schultz really was on the ice for more close-in chances than his peers. Finding his rate of shots against within 25 feet helps to answer that question (namely that, no, he’s not).

By using this past data, we’ll be able to build expected goal probabilities for each shot taken on the ice, no matter how far away they are. Then we’ll be able to add these all up to show the expected number of goals a player is on the ice for and against based on shot distances. I’ll show the results of that in the next post.

## 5 Comments

Is this going to include shot type, or is that too much work?

You could definitely take a look at shot types, as it’s in the data. Some people have and have found that different shot types have different rates of going in by distance. For now I think I’ll keep this pretty general, as it makes it easier to calculate and communicate. And I’d think that in certain instances, shot type is dependent on player’s choice based on what he thinks his best chance to score is or demanded by what hand he plays (backhand, etc). In any case, a close chance is a close chance.

Fair enough. I’ve been interested in seeing which players could drive play at an above average rate based on shot type and shot distance for a while. Though, I’m guessing this would be harder to determine on the defensive end since there is a greater team effect in that aspect, but I think it’s worth trying out.

Very interesting, basically from a tactics standpoint if you can keep your opponent from shooting within a 25 foot radius and conversely shoot within in that same radius (frequently) you can greatly increase your likelyhood of outscoring your opponent.

It would be interesting to look at goalie performance from this perspective ie. is a goalie with a high save percentage good or is his team more effective at preventing shots within that 25 foot radius. An example would Jason Labarbara who had a good EV save percentage and has fallen off the rails. Was he any good or was Phoenix good a preventing shots in the 25 foot area providing added likelyhood of success.

Very interesting…

I’d like to see you combine shots quantity (measured by corsi or fenwick) with a variable times avg shot distance in order to come up with a tweaked Corsi/Fenwick that predicts who won a game slightly better than does regular Corsi/Fenwick

## 2 Trackbacks

[…] time and time again. There’s no doubt about it. The most recent example is the distance analysis by Michael Parkatti, Different shots have different probabilities of going in and there are […]

[…] illustrated time and time again. There’s no doubt about it. The most recent example is the distance analysis by Michael Parkatti, Different shots have different probabilities of going in and there are […]