>Even with the increased use of expected-goals models, NHL goalies are still largely judged by their save percentage, but not all saves are created equal, so let’s add some context to those numbers.
>Shot quality is factored into the formula for goals saved above expected, which makes it a much better gauge of goaltending performance. It’s still not perfect, but it’s a major upgrade.
>Some goalies face a higher portion of easier shots from the outside of the offensive zone. Others face a consistent barrage of grade-A chances. So, let’s look at it this way: Which teams are allowing more of the outside shots — which should inflate their goalie’s save percentage — and which are allowing the most difficult shots that could be making their goalie’s save percentage look worse than it should?
>To quantify this, we came up with a simple ratio of low-danger shots to high-danger shots. If a team allows two low-danger shots for every one high-danger shot, they would have a ratio of 2.0. The higher the number, the easier the workload for the goalies.
>First, let’s quickly define high- and low-danger shots, as tracked by Natural Stat Trick, which we’ll be using for all the high- and low-danger data. Low-danger shots are any shot on goal with an assigned value of one (we’ll get to how that’s calculated next). High-danger shots are any shot on goal with a value of 3 or higher.
>To calculate the value of each shot, the biggest variable is shot location. Any shots from the crease or directly in front of it are given a value of 3, which automatically makes them high-danger shots. Any shots from the rest of the slot area — commonly referred to as the house — are given a value of 2, and all shots from the sides of the zone, along the walls, or the point are given a value of 1.
>You then add one to a shot’s value if the attempt is considered a rush shot or a rebound. A rebound is defined as any attempt made within three seconds of another blocked, missed or saved attempt without a stoppage in play in between. A rush shot is any attempt within four seconds of any event in the neutral or defensive zone without a stoppage in play in between.
>The league-average save percentage this season on all shots is .904. That’s a good baseline to start with. Now, the average save percentage on shots classified as “high-danger” is .806, while the average on low-danger shots is .961. With that massive gap between those numbers, you can see how easily a goalie’s overall save percentage can be skewed by the amount of each type of shot he faces.
>This exercise isn’t to determine which teams are playing the best defense, or allowing the fewest scoring chances. Those figures are easy to find. It’s to find which teams are inflating — or deflating — their goalie’s save percentage.
>Allowing a high volume of low-danger shots may not necessarily make a team’s defense better, but it increases a goalie’s save percentage and allows them to feel the puck to find a rhythm. Ask any goalie, and they’ll tell you a few low-danger shots against is ideal to start a game.
>Things have not gone as planned in Buffalo this season, and the team defense hasn’t made things easy on the goalies. Rookie Devon Levi has faced shots from an average distance of 32.35 feet, the second-closest of any goalie in the NHL.
>The Sabres don’t allow a particularly high number of shots. They rank 10th with only 29.1 against per 60 minutes. However, they give up the sixth-most high-danger shots per game, making it very tough for Levi and Ukko-Pekka Luukkonen to put up solid statistics. That Luukkonen has managed an above-average save percentage of .910 is a testament to how well he’s played.
1 Comment
>Even with the increased use of expected-goals models, NHL goalies are still largely judged by their save percentage, but not all saves are created equal, so let’s add some context to those numbers.
>Shot quality is factored into the formula for goals saved above expected, which makes it a much better gauge of goaltending performance. It’s still not perfect, but it’s a major upgrade.
>Some goalies face a higher portion of easier shots from the outside of the offensive zone. Others face a consistent barrage of grade-A chances. So, let’s look at it this way: Which teams are allowing more of the outside shots — which should inflate their goalie’s save percentage — and which are allowing the most difficult shots that could be making their goalie’s save percentage look worse than it should?
>To quantify this, we came up with a simple ratio of low-danger shots to high-danger shots. If a team allows two low-danger shots for every one high-danger shot, they would have a ratio of 2.0. The higher the number, the easier the workload for the goalies.
>First, let’s quickly define high- and low-danger shots, as tracked by Natural Stat Trick, which we’ll be using for all the high- and low-danger data. Low-danger shots are any shot on goal with an assigned value of one (we’ll get to how that’s calculated next). High-danger shots are any shot on goal with a value of 3 or higher.
>To calculate the value of each shot, the biggest variable is shot location. Any shots from the crease or directly in front of it are given a value of 3, which automatically makes them high-danger shots. Any shots from the rest of the slot area — commonly referred to as the house — are given a value of 2, and all shots from the sides of the zone, along the walls, or the point are given a value of 1.
>You then add one to a shot’s value if the attempt is considered a rush shot or a rebound. A rebound is defined as any attempt made within three seconds of another blocked, missed or saved attempt without a stoppage in play in between. A rush shot is any attempt within four seconds of any event in the neutral or defensive zone without a stoppage in play in between.
>The league-average save percentage this season on all shots is .904. That’s a good baseline to start with. Now, the average save percentage on shots classified as “high-danger” is .806, while the average on low-danger shots is .961. With that massive gap between those numbers, you can see how easily a goalie’s overall save percentage can be skewed by the amount of each type of shot he faces.
>This exercise isn’t to determine which teams are playing the best defense, or allowing the fewest scoring chances. Those figures are easy to find. It’s to find which teams are inflating — or deflating — their goalie’s save percentage.
>Allowing a high volume of low-danger shots may not necessarily make a team’s defense better, but it increases a goalie’s save percentage and allows them to feel the puck to find a rhythm. Ask any goalie, and they’ll tell you a few low-danger shots against is ideal to start a game.
>Teams hurting their goalies:
>4th: 4. Buffalo Sabres
Low-danger shots allowed: 603
High-danger shots allowed: 423
Ratio: 1.43
>Things have not gone as planned in Buffalo this season, and the team defense hasn’t made things easy on the goalies. Rookie Devon Levi has faced shots from an average distance of 32.35 feet, the second-closest of any goalie in the NHL.
>The Sabres don’t allow a particularly high number of shots. They rank 10th with only 29.1 against per 60 minutes. However, they give up the sixth-most high-danger shots per game, making it very tough for Levi and Ukko-Pekka Luukkonen to put up solid statistics. That Luukkonen has managed an above-average save percentage of .910 is a testament to how well he’s played.