Measuring defensemen effectiveness by opposition’s shot location
Around these parts, we spend a lot of time using #fancystats to supplement what we see on the ice from players. Generally speaking, we focus on puck possession metrics, using external factors like zone starts and quality of competition/teammates to pad the analysis. For all the merits that Corsi/Fenwick have, they don’t measure shot quality. Shot location isn’t something that is repeatable on offense(lots of analysis done to prove this), so there isn’t much space spent on examining further. It doesn’t correlate to wins.
However shot location can show us trends. We can use it at the team and individual levels to measure offensive styles –going to the net or keeping to the perimeter– and defensive effectiveness. On offense, you want to have more shots come from the better real estate (the slot, closer to the net). On defense, you want more shots from the perimeter. Hockey 101. War-on-ice, an indispensable resource, has given us a nice graph to show this as well.
WOI calls it a hextally chart, which measures shot locations relative to the rest of the league. Below is Dan Girardi’s chart:
The left side is how the team performs on offense with Girardi on the ice. You see all blue areas for the most part (you want red, which means above league average). That means when Girardi is on the ice, they aren’t getting league average levels of shots (in this case, 0 is the number that represents league average). The Rangers are slightly above average with shots from the slot, likely deflections of point shots. But offensively, the team struggles to get shot attempts (represented by all the blue).
Defensively –the right chart– is where you want to invert your expectations offensively. You want them all to be blue, not red, which is what we see from Girardi here. In particular, teams are having a pretty easy time getting shots from the slot and the high slot. Those aren’t promising charts.
We can use these charts to measure each defensive player. Next up is Ryan McDonagh:
McDonagh certainly drives offense a lot more efficiently than Girardi, as noted by the lack of blue dots on the left side chart and a significant uptick in slot attempts. However, we see that his defensive abilities to limit shots to the outside is lacking so far this season. It’s early, so we expect this to correct itself for someone like McDonagh, but it really illustrates the rough start to the year. Someone of McDonagh’s caliber should have more blue dots defensively.
Marc Staal is another guy having a rough start to the year:
Staal’s simply not driving offense, as shot attempts are way below league average all over the ice. On the bright side, you can see that he’s doing a somewhat decent job at keeping slot chances at a minimum (in the negatives there, which is good). The team does appear to give up a lot of shots from the center point when he’s on the ice though.
Kevin Klein, who has had a tremendous start to the season, has been one of the more effective players at launching shots from the point:
Defensively, he’s been one of the better guys at limiting shots from high-priced real estate. He’s been pretty good at limiting shot attempts against, as there aren’t many red dots on this chart.
And finally, John Moore:
Moore is the only Ranger who seems to have more shot attempts, relative to the rest of the league, that are all above the league average per zone. Considering the type of defensemen he is, it’s relatively expected. He’s not all that good at suppressing shot attempts from the slot though.
The hextallys are a good start to showing where defensemen are allowing shots. However, it doesn’t tell the full story, since it doesn’t include the types of shots (rushes, rebounds, etc), situation, gap control, and a myriad of other factors. It’s just raw shot attempt data. However, it’s interesting to see which defensemen are better at limiting chances to the outside. As much as we want our defensemen to have low Corsi-Against ratings, we want them to keep the shots to the outside as well.