What are PHF priorities this offseason? — Mikael Nahabedian talks SARAH, women’s hockey analytics — Must-click women’s hockey links

The IX: Hockey Friday with Anne Tokarski, April 1, 2022

Welcome back to Hockey Friday. The PHF crowned a new champion on Monday — actually, scratch that. They crowned an old champion in the Boston Pride, who became the first team to win the Isobel Cup in the PHF era of professional women’s hockey in North America and the first PHF/NWHL team to win three championships in the league’s history. Last season, the Pride became the first team to win two championships…so it seems only fitting that they continue setting some records.

In other news, Ian Kennedy of The Hockey News and Dan Rice of The Ice Garden reported on Twitter that the PHF Players’ Association is close to naming a new director of the PA in Nicole Corriero…and I figured this appointment meant it was high time to tackle some of the other matters of business the PHF ought to address before season eight starts up in the fall.

First up: single-elimination postseason. Angelica Rodriguez wrote an excellent column on this very same topic that can be found here, and I’d pretty much echo and endorse everything she says in that piece. I think it’s crucial that, if the resources are available, the PHF expand their postseason format to a best-of-three or best-of-five series. If not possible for the entirety of the playoffs, then the championship, at the very least, should be awarded at the culmination of a series. We’ve seen it before in women’s hockey with a single unlucky bounce deciding the fate of a team…but it should be the talent and hard work of the athletes that decides who wins a championship, not a fluke.

(Disclaimer: This is not a commentary on this year’s championship or that of any years previously!)

Next, it’s obvious we want players to be paid a living wage playing professional hockey. Until we get to that point, I don’t think we see players relocating to play far from their hometowns or where they work, because it’s simply not feasible. But there needs to be some way to ensure that all of the talent isn’t centralized in one location and ensure that there are enough available roster spots for players who want to compete. The PHF confirmed that this offseason will see a double expansion with Montréal and another city to be named, so the availability of roster spots isn’t of too much concern to me right now.

Finally, I think it’s really, really important that the league needs to prioritize accurate stats and data. Too often we see incorrect box scores posted after the games, and it’s become a point of contention for a lot of people in trying to provide accurate coverage of games. When someone can’t be physically present at the game the box scores are wrong, how can you expect adequate coverage?


The Midge Purce Promo!

It’s NWSL season, meaning it is Midge Purce season! Celebrate #23 with 23 percent off a subscription to The IX, where every Monday is Soccer Monday and every day we cover six different women’s sports! Subscribe now and join us for just $46.20 for your first year. It’s the women’s sports media network we all wished for, and now it’s here!


This week in women’s hockey

As women’s hockey grows, Michigan’s absence is glaring (Michigan Daily)

Minnesota’s Taylor Heise named Patty Kazmaier Award winner (The Ice Garden)

RPI’s Price finds volunteering a way to ‘take back control over my life’ (USCHO)

PWHPA set to play first showcase in Montréal this weekend (NHL.com)

The PHF has a busy offseason ahead (The Hockey News)

Ohio State’s Nadine Muzerall named USCHO Coach of the Year (USCHO)


Five at The IX: Mikael Nahabedian

Mikael Nahabedian (@hunterofstats) is a data analyst who has worked with the McGill Martlets hockey program in Montréal and the SDHL’s AIK Hockey.

Question: Tell me a little bit about how you got started in women’s hockey analytics and how you got to where you are now — your background, your current projects, etc.

Mikael Nahabedian: I started working in women’s hockey analytics 4 seasons ago when I got the chance to join the McGill Martlets as a stats tracker. Over time, I’ve grown into my role with the Martlets and led our video & analytics team, in addition to helping with our scouting process this past season. Over the years, I’ve also worked in various capacities with professional club teams in the NWHL (now PHF) and SDHL, as well as some up-and-coming national teams in Europe.

Since the beginning of my career, I’ve had the chance to publish several hockey analytics research projects in the public sphere. These projects have been related to scouting (N-WHKYe), tactics (Possession Value Added) and player development (SARAH).

After spending the last 4 seasons with the Martlets, I’m currently open to exploring new opportunities in hockey analytics.

Q: You recently debuted SARAH with a handful of other data analysts. Tell me a little bit about the project and what you hope to accomplish with your model.

Nahabedian: We developed SARAH with an awesome team comprised of coaches, players and data analysts with Noah Chaikof, Katia Clément-Heydra, Carleen Markey, Adam Pilotte and Mairead Shaw.

The SARAH model serves a dual purpose: to link advanced stats to traditional video scouting and streamline the process of evaluating the developmental progress of a players’ habits. Essentially what we did was first look at the various national women’s hockey teams and identify the set of “habits” a player regularly executes (i.e., edgework, catching the puck in the hip pocket, pass placement, etc).

Combining the dataset of players’ habits with a set of players’ microstats (entries via pass/stickhandling, exits via stickhandling/pass, accurate/inaccurate passes, etc.), we developed SARAH to accurately predict if a player possesses a certain habit based on their set of microstats. Combining this with an estimate of how frequently players used each habit, we created an interactive Player Development Matrix to visualize different player’s habits.

Q: You were able to use comprehensive data from the IIHF to give your project an interactive product. Was it at all rewarding to be able to compare the skill levels of players across all countries rather than focusing on the typical US and Canadian skaters?

Nahabedian: Yes, it was very interesting to get the opportunity to develop a tool that could evaluate players across all countries in the World Group. Providing a tool that could potentially help other teams develop their players was also one of the objectives of this project.  In turn, the long-run development of players at the micro-habit level could allow these other teams to become more competitive when facing Team Canada and Team USA.

In addition to that, getting a chance to track over 250 players from all around the world gave us the opportunity to discover new players with interesting skill sets. Getting a comprehensive view at these 262 skaters was very insightful. We hope that this project can introduce some of these hidden gems to women’s hockey fans.

Q: We often talk about how there’s somewhat of a lack of available (and public) data for women’s hockey players’ on-ice habits in professional organizations like the PHF and PWHPA. How do you hope to see that change in the future?

Nahabedian: It is an exciting time for women’s hockey analytics as new faces such An Nguyen, Ben Howell & Saiem Gilani are joining this emerging field. Public analytics work is important to set strong foundations and there are a lot of opportunities for analysts to get started in hockey analytics by working on some cool women’s hockey data projects.  With the recent launch of the Big Data Cup organized by Stathletes, which is centered around women’s hockey data this year, I think we’re going to see more and more cool projects being published in the coming months.

I hope that in a near future more and more young analysts build on the work of pioneers in the field like Alyssa Longmuir, Carleen Markey, Mike Murphy, Spencer Fascetta, Jack Han and Nayan Patel. By setting these strong foundations, more and more professional organizations will eventually want to invest in an accessible product that can provide competitive advantage through actionable insights.

Q: You collaborated with former Les Canadiennes de Montréal player Katia Clément-Heydra on SARAH. What has it been like to see women’s hockey players and coaches get more involved in statistical and data analysis?

Nahabedian: It has been great to see more women’s hockey players and coaches get involved in using data to, respectively, improve their skills and drive some of their decisions. From my experience, most women’s hockey players are very receptive to analytical feedback. However, with players, I prefer to embed analytical concepts within video analysis to not burden them with complex modelling interpretations. With players, I rather illustrate through video examples what they need to do well to be successful on the ice.

For coaches, understanding how to use the numbers provided by analysts is crucial as analytics is taking an ever-growing importance in hockey. Linking analytical insights to hockey concepts is the way to optimize decision-making in a modern organization. Speaking of Katia, she exemplifies this last point very well as a tremendous hockey mind who has a unique way of integrating analytical concepts into her coaching. Thanks to this, I have no doubt that she will have a very long and very successful career at the highest levels of hockey.


Mondays: Soccer
By: Annie Peterson, @AnnieMPeterson, AP Women’s Soccer
Tuesdays: Tennis
By: Joey Dillon, @JoeyDillon, Freelance Tennis Writer
Wednesdays: Basketball
By: Howard Megdal, @HowardMegdal, The Next
Thursdays: Golf
By: Addie Parker, @addie_parker, The IX
Fridays: Hockey
By: Anne Tokarski, @annetokarski, The Ice Garden
Saturdays: Gymnastics
By: Lela Moore, @runlelarun, Freelance Writer

Written by Anne Tokarski