Fan Analytics at the NHL | Jamie Shive | Data Science Hangout
ADD THE DATA SCIENCE HANGOUT TO YOUR CALENDAR HERE: https://pos.it/dsh - All are welcome! We'd love to see you!
This week's guest was Jamie Shive, Director of Data Science at the National Hockey League (NHL)!
Some topics covered in this week's Hangout were fan analytics (fanalytics? 🤔) across a sports league, building fan persona segmentation with clustering, the evolution of fandom from team loyalty to player loyalty in the age of social media, and building a data science team with diverse backgrounds.
One community member asked: Do you have any advice on how to map customer personas segments so they tell an actual story to the business you're working with?
Jamie's paraphrased answer: Being a fan herself helped her tell if the personas felt right, but if you don't have that context, tap a key stakeholder from each function of the organization to look at what you've produced and confirm whether they recognize those fans. The bigger lesson was to build the personas around digital behavior engagement alone rather than layering in demographics and psychographics up front. Mixing all of that together produced "giant clumps of nothing" that didn't distinguish the people actually showing up. By keeping behavior as the defining feature and layering demographics or psychographics on top afterward, the personas stayed distinct and could be shared with every vertical of the organization.
Resources mentioned in the video and chat:
🔗 MoneyPuck Hockey Data → https://moneypuck.com/data.htm
🔗 "The Worst Lead in Hockey" on Zajichek Stats blog post → https://www.zajichekstats.com/post/the-worst-lead-in-hockey/
🔗 nflverse – Sports data R packages → https://nflverse.nflverse.com/
🔗 Matt Dray's Splendid R Games list → https://matt-dray.github.io/splendid-r-games/
🔗 Hex Memory game (Shiny Gallery) → https://shiny.posit.co/r/gallery/miscellaneous/hex-memory/
🔗 shinydoom – DOOM built in Shiny → https://github.com/pachadotdev/shinydoom
🔗 Nichecraft by Lynda C. Falkenstein, → book on building niches and brand fanatics for businesses https://www.amazon.com/Nichecraft-Specialness-Business-Corner-Customers/dp/0962574724
â–º Subscribe to Our Channel Here: https://bit.ly/2TzgcOu
Follow Us Here:
Website: https://www.posit.co
Hangout: https://pos.it/dsh
The Lab: https://pos.it/dslab
LinkedIn: https://www.linkedin.com/company/posit-software
Bluesky: https://bsky.app/profile/posit.co
Thanks for hanging out with us! 💛
Timestamps:
00:00 Introduction
05:54 "Could you give us an example so that we understand what type of tools you use, what type of analytics you do, and how data helps solve fan problems or improve fan experience across the league?"
09:02 "What kind of tools do you use?"
10:11 "One of the concerns with working in sports analytics is night and weekend work. Have you found that to be the case for you in fan analytics?"
12:48 "What is unique to sports analytics, and in your case fan analytics, compared to other domains? And what role do you think things like Polymarket are playing in the fan analytics realm?"
17:47 "What's the difference between working for a league versus working for a specific team from an analytics perspective?"
20:21 "How do you support teams that have their own analytics, and do you do a lot of supporting teams when working with their fan analytics?"
25:14 "Do you have any advice on how to map those fan persona segments you've created to tell an actual story to the business you're working with?"
29:43 "Do the segments change over time? Has it been long enough for you to recheck them, and are you going to recheck?"
32:52 "What is the most interesting insight about fans that your data has discovered, besides people in the game zone not buying merch but opening emails?"
37:25 "Are you noticing the younger generations, like Gen Z and Gen Alpha, acting differently than older sports fans?"
41:36 "Do you have people with a psych background or other forms of data wrangling backgrounds? And do you think sports analytics as a whole is diverse enough to cater to all of these different needs, or does it need more people from different backgrounds and foundations?"
44:18 "Do you have any advice for a new budding data scientist who wants to get into hockey or sports analytics? Any datasets, projects, or techniques that you would like to see in a portfolio?"
47:29 "Has the recent rise in new NHL fans from the popularity of books or shows like Heated Rivalry and Off Campus affected your analyses?"
51:16 "Do you have a piece of career advice, or something you wish you knew before you went into data science that would have helped you out?"
This week's guest was Jamie Shive, Director of Data Science at the National Hockey League (NHL)!
Some topics covered in this week's Hangout were fan analytics (fanalytics? 🤔) across a sports league, building fan persona segmentation with clustering, the evolution of fandom from team loyalty to player loyalty in the age of social media, and building a data science team with diverse backgrounds.
One community member asked: Do you have any advice on how to map customer personas segments so they tell an actual story to the business you're working with?
Jamie's paraphrased answer: Being a fan herself helped her tell if the personas felt right, but if you don't have that context, tap a key stakeholder from each function of the organization to look at what you've produced and confirm whether they recognize those fans. The bigger lesson was to build the personas around digital behavior engagement alone rather than layering in demographics and psychographics up front. Mixing all of that together produced "giant clumps of nothing" that didn't distinguish the people actually showing up. By keeping behavior as the defining feature and layering demographics or psychographics on top afterward, the personas stayed distinct and could be shared with every vertical of the organization.
Resources mentioned in the video and chat:
🔗 MoneyPuck Hockey Data → https://moneypuck.com/data.htm
🔗 "The Worst Lead in Hockey" on Zajichek Stats blog post → https://www.zajichekstats.com/post/the-worst-lead-in-hockey/
🔗 nflverse – Sports data R packages → https://nflverse.nflverse.com/
🔗 Matt Dray's Splendid R Games list → https://matt-dray.github.io/splendid-r-games/
🔗 Hex Memory game (Shiny Gallery) → https://shiny.posit.co/r/gallery/miscellaneous/hex-memory/
🔗 shinydoom – DOOM built in Shiny → https://github.com/pachadotdev/shinydoom
🔗 Nichecraft by Lynda C. Falkenstein, → book on building niches and brand fanatics for businesses https://www.amazon.com/Nichecraft-Specialness-Business-Corner-Customers/dp/0962574724
â–º Subscribe to Our Channel Here: https://bit.ly/2TzgcOu
Follow Us Here:
Website: https://www.posit.co
Hangout: https://pos.it/dsh
The Lab: https://pos.it/dslab
LinkedIn: https://www.linkedin.com/company/posit-software
Bluesky: https://bsky.app/profile/posit.co
Thanks for hanging out with us! 💛
Timestamps:
00:00 Introduction
05:54 "Could you give us an example so that we understand what type of tools you use, what type of analytics you do, and how data helps solve fan problems or improve fan experience across the league?"
09:02 "What kind of tools do you use?"
10:11 "One of the concerns with working in sports analytics is night and weekend work. Have you found that to be the case for you in fan analytics?"
12:48 "What is unique to sports analytics, and in your case fan analytics, compared to other domains? And what role do you think things like Polymarket are playing in the fan analytics realm?"
17:47 "What's the difference between working for a league versus working for a specific team from an analytics perspective?"
20:21 "How do you support teams that have their own analytics, and do you do a lot of supporting teams when working with their fan analytics?"
25:14 "Do you have any advice on how to map those fan persona segments you've created to tell an actual story to the business you're working with?"
29:43 "Do the segments change over time? Has it been long enough for you to recheck them, and are you going to recheck?"
32:52 "What is the most interesting insight about fans that your data has discovered, besides people in the game zone not buying merch but opening emails?"
37:25 "Are you noticing the younger generations, like Gen Z and Gen Alpha, acting differently than older sports fans?"
41:36 "Do you have people with a psych background or other forms of data wrangling backgrounds? And do you think sports analytics as a whole is diverse enough to cater to all of these different needs, or does it need more people from different backgrounds and foundations?"
44:18 "Do you have any advice for a new budding data scientist who wants to get into hockey or sports analytics? Any datasets, projects, or techniques that you would like to see in a portfolio?"
47:29 "Has the recent rise in new NHL fans from the popularity of books or shows like Heated Rivalry and Off Campus affected your analyses?"
51:16 "Do you have a piece of career advice, or something you wish you knew before you went into data science that would have helped you out?"