By the Numbers: Investigating Goaltending Storylines

Cam Ward made his first return to PNC Arena with the Blackhawks on Monday night for a game in which both he and Scott Darling played very well against their former teams. Given the irony of that situation, I thought it would be a fitting week to assess the goaltending condition in Carolina and across the league as a whole.

The age old Canes narrative has been that the team severely lacks top end scoring and decent goaltending talent. As the rebuild has come to a head over the past few seasons and offensive depth has begun to take hold in Raleigh, the narrative has slowly shifted to something like this:

I honestly don’t disagree with this take — he’s absolutely right that goaltending statistics coming out of Carolina have been well below average since even before the Bill Peters tenure began in 2014. But I’m just not entirely convinced that it tells the whole story.

I have a hunch about this team and others across the league that significantly outshoot and outchance their opponents on a night-to-night basis, and I’m hypothesizing that those types of systems foster what appears to be below average goaltending. I’m not sure why it is, but if I had to guess I’d start with a statement that Scott Darling made in his post-game interview after the deflating Saturday night loss to Detroit. He mentioned that in Carolina there are a lot of “dry periods.” He elaborated that most of his action came in the first 30 minutes of the game, and that in the second half of the game he was just trying to stay sharp despite only seeing a few shots.

I got to thinking that it must be tough on a goaltender to have long periods of inactivity and then get immediately thrown into the fire in the form of a high danger scoring chance. The Hurricanes outchance opponents by a wider margin than any almost every team in the league, and also have one of the lowest 5 on 5 team save percentages.

Nothing new, right? Same old Canes. But the interesting thing is that the Hurricanes aren’t the only team with these types of problems this season. The only other club with a higher Scoring Chances For Percentage than Carolina’s 57.33% is the San Jose Sharks, who are boasting a 57.45%. Their 5 on 5 Save Percentage is a brutal .890, the second worst in the league.

The worst save percentage in the NHL belongs to the Vegas Golden Knights at .888%. Their SCF%? 56.43%, good for third best in the NHL behind Carolina and San Jose. In the same vein, Montreal has the fifth best SCF% in the league this season, but the third worst 5 on 5 SV%.

The excuse presented for poor Hurricanes goaltending over the past few years has been a lack of a true NHL caliber starting goaltender. But what’s the excuse for teams like Vegas, San Jose, and Montreal? Each team has a traditionally solid number one goaltender, but have struggled this season to post league average statistics in that category despite outchancing their opponents handily.

The above graph shows Scoring Chances For Percentage on the x-axis and 5 on 5 Save Percentage on the y-axis. Remember that I hypothesized that teams with higher SCF% would have lower 5 on 5 SV%. The graph shows a very small trend in that direction, but its not nearly enough for me to reasonably argue causation.

Despite no correlation in the middle of the pack, it seems like teams on either extreme of SCF%, but particularly on the upper extreme, conform to the idea of these two data points being related. The average save percentage today in the NHL is about .919%, and the top five teams in SCF% all post even strength save percentages below average, sometimes by a very large margin. But what if we compared SCF% with a more telling goaltending metric in High Danger Save Percentage?

Just as before, there doesn’t seem to be much of a correlation here. But it does seem like those goalies who play for teams in the upper echelon of SCF% seem to perform below average in High Danger Save Percentage. In this case, league average is about 83%, and the best six teams in SCF% are all coming in below that figure.

In conclusion, the relationship between goaltending performance and team competitiveness is complicated. But there appears to be a connection between heavily outchancing your opponent and below average goaltending. Maybe a system like Rod Brind’Amour teaches in which goaltenders see less action and fewer shots negatively impacts their play. Perhaps teams who seriously outshoot and outchance opponents are more susceptible to huge defensive breakdowns when things go the other way. Or maybe the problem is something entirely different. The answer will require more digging.