Long-time listener and first-time caller Sunny Mehta explores an interesting topic over at Irreverent Oiler Fans. Whether you’re a believer in shot quality or not, it’s impossible to disagree with Mehta when he notes:
“We observe a large difference in shooting percentages between
forwards and defensemen.”
Overall, he finds that defensemen have a 4.1% shooting percentage, while forwards are at 9.6%. This isn’t surprising – forwards both have more offensive skill and play in locations on the ice where they are more likely to score.
It’s easy to see the difference in shot location. I looked at all 5-on-5 shots from 2001-2009, and removed rebounds:
The goal is located at (0,0) on these charts – you can clearly see the shape of the crease in the shot diagram for forwards. Intuitively, this is what we expect: defensemen shoot from the point while forwards shoot from the slot or from the tops of the faceoff circles. So it’s no surprise that forwards have a higher shooting percentage – 8.58% in my sample – than defensemen – 4.01%.
What if we separated shot location from the rest of our observations? Let’s assume that instead of having defensemen shooting from where they did, we had forwards shooting from these spots. We can do this by taking forwards’ shooting percentage and weighting it by the locations that defensemen shot from. We need to do two things to figure this out. First, we determine shooting percentage for forwards at every x-y location on the ice – let's refer to that as SPct_Fwd(x,y). We also know the number of shots defensemen took from every spot on the ice – let’s call that Shots_ Def(x,y).
If we now multiply the SPct_Fwd matrix by the Shots_Def matrix, we get an estimate of the number of goals that forwards would have scored had they been restricted to shooting from the spots that defensemen did. The overall shooting percentage for that entire set of shots is 4.59%, compared to the 4.01% shooting percentage that defensemen actually had.
So, given the same opportunities as defensemen, forwards had a shooting percentage that was 14% higher. There could be a lot of things wrapped up in this 14% - more or fewer screens, or the difference between shot types: for example, a winger taking a shot on an odd-man rush versus a defenseman taking a shot off of a face-off. Fundamentally, forwards are selected for their offensive ability, and the unsurprising outcome here is that they do in fact have better offensive abilities, which includes taking higher quality shots than their defensemen counterparts.
Gabriel Desjardins is an author of Puck Prospectus and runs the statistical hockey site behindthenet.ca. You can contact him at info at behindthenet.ca.