State-of-the-Stats NWHL Edition
There was a call for those working in advanced stats over the summer to start working on analytics for the NWHL and CWHL. Now I’m not going to delve into Corsi or Fenwick or anything player related. I like to focus on everything goaltending related and this time will be no different. Personally I’m just starting into the deep sea that is analytics/advanced stats. There will not be a detailed story for each goalie in the NWHL or CWHL just yet. That will take time. In the meantime though I have started on integrating Quality Starts % into each respective league.
Quick overview on Quality Starts %. A goalie is rewarded a Quality Start if they post a Save Percentage in a game over the league’s average save percentage or allow 2 or less along with posting a SV% over the replacement save percentage. A Quality Start means that your goalie has given you a 77.5% chance of winning the game. When a goalie posts a QS% of over 0.60 then they have reached an elite level. This stat was created by Rob Vollman.
When you combine QS% with Points % you can get a picture of how much support a goalie gets from their team. For example in the CWHL it was obvious that Boston Blades goalie Genevieve Lacasse was under siege game after game. The question though has been just how much weight was Lacasse pulling on the Blades? And then there is the Goalie of the Year award. Labonte definitely deserved it; however how close was Erica Howe to unseating her? Looking at PTS%/QS% gives us a look at part of the story for each starting goalie in the CWHL.
Now the main focus of the article is the NWHL. Looking at the PTS%/QS% chart for the NWHL tells a story of how each starting goalie in the NWHL was consistently good at least and at best elite. No starting goalie was under 0.50 QS%. Three out of the four goalies ended up over 0.60 which means elite level goaltending. And one goalie, Jamie Leonoff ended with a 1.00 QS%. Leonoff is the goalie that GM’s love; consistently great game in, game out. The Riveters are going to miss her presence in net.
If there is one knock against Jamie Leonoff it’s that she played the least amount of games among the four starters. I’m assuming it’s this reason that Leonoff was not given Goalie of the Year in the NWHL this past season. If we look at this like Brianne McLaughlin vs Brittany Ott for the NWHL Goalie of the Year we see one goalie who outplayed her team and another goalie that played well however her own team outplayed her. Keep in mind though that Brittany Ott still put up a 0.67 QS% so she definitely deserved GotY. Brianne McLaughlin though might be the underrated goalie of the NWHL.
Now I’d like to use QS% and combine it with other stats such as SA/60 to get a look at each goalies workload. However there’s a problem with the NWHL. The problem is that minutes for each goalie were not tracked properly. This primarily means that Goals Against Average and Shots Against per 60 are now useless. In fact the GAA on the official NWHL website is wrong. The calculations are correct as the numbers are available, it’s just that the numbers are wrong and that is because minutes played for goalies were improperly taken care of.
I even went through the numbers myself on my own excel spreadsheet to make sure that the NWHL wasn’t using their own numbers. It was possible that the NWHL actually had the correct minutes played for each goalie and just weren’t sharing that data because it would look weird compared to the actual game sheets. Unfortunately that is not the case. What we’re left with is inaccurate data which isn’t a deal breaker but it makes doing advanced stats difficult to do. I really hate giving up too but there is no possible way to find the accurate minutes played by each NWHL goalie. Even the live game sheets claim there were never goalie changes and the starting goalie played all 60 minutes.
The last gripe I have with the NWHL stat takers are the games played category. Each NWHL team plays 18 in a season. Looking at the NWHL website you’ll see some interesting games played numbers. For instance both Brittany Ott and Kelsie Fralick played 18 games for the Boston Pride. That wouldn’t be out of place if this was Atom hockey and the coach elected to give the goalies half the game each game. This isn’t Atom hockey though, this is pro hockey. Even if you didn’t watch the games it’s obvious that Ott and Fralick weren’t each playing half the game then switching.
This season though it looks as though the NWHL has caught the minutes played mistake. Good work by the NWHL and I’m excited to apply more stats to their goalies. It’s an exciting time and I hope that the talented goalies of the NWHL start to gain more recognition for their play. With the way the NWHL is right now it’s important to be able to put together stats that accurately describe the play of each goalie because of the differences in each team. Comparing raw SV% between Brittany Ott and Nana Fujimoto doesn’t show an accurate picture of their play so we need more of a breakdown which will hopefully be provided this season.