Goaltender Workload and Its Effect on Save Percentage
Carey Price has seen the 10th most shot attempts per game this season, and has 61.67 seconds to recover per shot attempt, on average. (Scott Slingsby photo)
With the advanced stat revolution taking the hockey world by storm, it’s easy to forget that some of the data collected was originally used to monitor the workload of goaltenders.
This week, the NHL finally unveiled a new website that allows users to look up shot attempts and unblocked shot attempts, along with a variety of other advanced statistics. Although these stats are in the forefront now, they have actually been around for a very long time.
Shot attempts were collected by Jim Corsi, who is currently the goalie coach of the St. Louis Blues, while he was coaching with the Buffalo Sabres in the early 00s. Instead of using the data collected as a proxy for puck possession, he was using the information to gauge how much action his goalies were seeing each game.
Every goalie knows that it doesn’t require a shot to expend energy. Simply having the puck in the offensive zone requires the goaltender to be ready, and shot attempts are a more accurate way to judge how busy a goaltender has been.
By examining shot attempts against, it’s easy to find out which goalies have been the busiest this season, and which goalies have not seen as much action.
To remove the bias of goaltenders that play more consistently, calculating the average length of time each goalie has to wait before seeing each shot attempt is a more fair and intriguing way to look at the numbers.
Let’s break it into three categories:
Heavy Workload
| Name | Team | Average Length Between Shot Attempts (In Seconds) |
|---|---|---|
| Michal Neuvirth | Buffalo Sabres | 52.15 |
| Jhonas Enroth | Dallas Stars | 53.49 |
| Calvin Pickard | Colorado Avalanche | 56.38 |
| Semyon Varlamov | Colorado Avalanche | 57.92 |
| Jonathan Bernier | Toronto Maple Leafs | 57.99 |
| Kari Lehtonen | Dallas Stars | 59.52 |
| James Reimer | Toronto Maple Leafs | 60.26 |
| Jonas Hiller | Calgary Flames | 60.32 |
| Devan Dubnyk | Minnesota Wild | 60.41 |
| Carey Price | Montreal Canadiens | 61.67 |
| Mike Smith | Arizona Coyotes | 62.38 |
| Viktor Fasth | Edmonton Oilers | 62.46 |
| Sergei Bobrovsky | Columbus Blue Jackets | 62.84 |
| Curtis McElhinney | Columbus Blue Jackets | 62.91 |
| Karri Ramo | Calgary Flames | 63.21 |
| Ray Emery | Philadelphia Flyers | 63.30 |
| Dustin Tokarski | Montreal Canadiens | 63.91 |
| Steve Mason | Philadelphia Flyers | 64.10 |
| Craig Anderson | Ottawa Senators | 64.17 |
| Antti Niemi | San Jose Sharks | 64.39 |
Predictably, the goaltenders high on the list play for teams that are terrible at possessing the puck. It’s interesting to see that it doesn’t really matter which goaltender is in net, teams that are terrible at possessing the puck will be terrible even when their backup is in.
Since it’s an average, it may not seem like much, but a ten second difference in rest time becomes more significant when you put it in the perspective of an entire 60 minute game. An extra ten seconds of resting in between each shot attempt can make a huge difference by the third period.
Now let’s check out the goalies that are seeing an average workload.
Average Workload
| Name | Team | Average Length Between Shot Attempts (In Seconds) |
|---|---|---|
| Justin Peters | Washington Capitals | 64.63 |
| Frederik Andersen | Anaheim Ducks | 64.76 |
| Keith Kinkaid | New Jersey Devils | 64.85 |
| Scott Darling | Chicago Blackhawks | 65.01 |
| Henrik Lundqvist | New York Rangers | 65.56 |
| Braden Holtby | Washington Capitals | 65.56 |
| Eddie Lack | Vancouver Canucks | 65.81 |
| Anton Khudobin | Carolina Hurricanes | 65.96 |
| Ryan Miller | Vancouver Canucks | 66.29 |
| Alex Stalock | San Jose Sharks | 66.45 |
| Andrei Vasilevskiy | Tampa Bay Lightning | 66.54 |
| Carter Hutton | Nashville Predators | 66.69 |
| Robin Lehner | Ottawa Senators | 66.77 |
| Thomas Greiss | Pittsburgh Penguins | 66.88 |
| Jaroslav Halak | New York Islanders | 67.03 |
| Ben Scrivens | Edmonton Oilers | 67.04 |
| Pekka Rinne | Nashville Predators | 67.21 |
| Tuukka Rask | Boston Bruins | 67.42 |
| Cam Talbot | New York Rangers | 67.43 |
| Darcy Kuemper | Minnesota Wild | 67.48 |
| Anders Lindback | Buffalo Sabres | 67.87 |
Some interesting names appear on that list. Ben Scrivens has seen far fewer shot attempts than most would initially have guessed. Pekka Rinne and Tuukka Rask are not nearly as busy as some of the other Vezina candidates. A few goaltenders that have struggled this season, like Anders Lindback and Robin Lehner, can’t pass it off as an unusually high workload – at least according to this data.
Which goalies have had an easy go of it this season so far?
Light Workload
| Name | Team | Average Length Between Shot Attempts (In Seconds) |
|---|---|---|
| Michael Hutchinson | Winnipeg Jets | 68.03 |
| Cam Ward | Carolina Hurricanes | 68.22 |
| Al Montoya | Florida Panthers | 68.72 |
| Roberto Luongo | Florida Panthers | 68.89 |
| Ondrej Pavelec | Winnipeg Jets | 68.94 |
| Antti Raanta | Chicago Blackhawks | 68.98 |
| Brian Elliott | St. Louis Blues | 68.99 |
| Cory Schneider | New Jersey Devils | 69.05 |
| Marc-Andre Fleury | Pittsburgh Penguins | 69.14 |
| Niklas Svedberg | Boston Bruins | 69.31 |
| Chad Johnson | New York Islanders | 69.40 |
| Jonathan Quick | Los Angeles Kings | 69.74 |
| Ben Bishop | Tampa Bay Lightning | 69.88 |
| Niklas Backstrom | Minnesota Wild | 70.03 |
| Jake Allen | St. Louis Blues | 70.12 |
| Corey Crawford | Chicago Blackhawks | 70.95 |
| Evgeni Nabokov | Tampa Bay Lightning | 74.78 |
| Petr Mrazek | Detroit Red Wings | 75.54 |
| Jimmy Howard | Detroit Red Wings | 77.39 |
| Martin Jones | Los Angeles Kings | 78.47 |
A lot of the information can be skewed by sample size, but all of the goaltenders on the list have played at least 400 minutes on the season by February 18th, 2015 – when this information was collected.
Goalies that play for strong possession teams unsurprisingly get a longer time to rest between shot attempts, but breaking it down this way gives a new perspective on how much action these particular goaltenders have to face each game, and shows which goaltenders deserve a bit more credit because of how much extra work they have put in.
The difference between the busiest goaltender in the league (Neuvirth) and the least busiest (Jones) is 26.32 seconds between shot attempts.
Does workload have an effect on a goaltender’s save percentage?
Common sense would say yes, but it yields some interesting results when this data placed on a chart. The busiest goaltenders are on the left side, and the least busiest are on the right. The red line is the league’s average save percentage:
Click to enlarge
There seems to be just as many goalies struggling to stay above the league average on the right side of the chart as the left! Shot attempt quantity doesn’t seem to be as big of a negative factor on save percentage as originally thought.
Shot quality is not taken into account because of its difficulty to quantify, and that may have the biggest effect on save percentage of all.
Although the save percentage numbers may not show it, workload is still something that NHL goaltenders have to deal with and manage every day. When a goalie faces an absurd number of shot attempts against, it is probably a good idea to give him a rest.
Teams will judge workload by using a variety of different statistics, but shot attempts continues to be one of the best ways of fully understanding how much strain a single game puts on a goaltender’s body.
Great piece. I think this effectively proves that shot quality is more important than quantity. Equally, it seems that far more intangibles are in play, such as some goalies having better conditioning, others being able to focus without work for longer and a host of others. The chart looks relatively even because we’re seeing all of these variables evening out and canceling each other depending on the player and team in front.
Thanks for the interesting article. I have always found that some goalies just need the work to be effective. Playing for a team where you don’t see a lot of action is often more difficult then playing for the team where you are busy. It really depends on the individual and therefore I think the results are consistent.
Fully agree Brian. There seems to me to be a sweet spot for each goalie between too much and not enough work, both within a game and on a day/week basis (i.e. sessions/week, days in a row, morning/evening, etc.). And as the article alludes to, we do a lot even when there are no shots, so attempts, blocks, and misses all come into play. And did anyone mention rebounds, flurries, etc.? Wow, lots of considerations….
I too find this to be an interesting article. At first I thought there would’ve been at least been some relationship between workload and save percentage. Very interesting how so many factors are in play, therefore making it somewhat consistent, as Brian mentioned.
Thanks for sharing this data. My son is on 2 very weak teams this season. His work load is tremendous to say the least. He is the main goalie for both his U16 team and High School team. He doesn’t complain. He says that it would be boring to just stand and wait the whole game. He says he just prays that God gives him the vision to see the puck and the ability to give his team a fighting chance to win.