Building a data-driven talent pipeline for community cricket
Cricket NSW and the UTS Data Science Institute collaborated on the Player Journey Intelligence Project, utilizing seven years of cricket data to develop clustering and machine learning models that help us better understand the pathways players take through their cricketing lives.
Image source: Adobe Stock
Tomorrow’s cricket superstars might be playing community cricket now, but getting them from junior blaster to master blaster means keeping them excited about the game.
Understanding the how and why of player retention is therefore a key strategic area for state cricket organisations seeking to build and maintain their talent pipelines. At Cricket NSW, unlocking the value of player data was the first step towards success.
In 2022, Cricket NSW and the UTS Data Science Institute (DSI) launched the Player Journey Intelligence Project to better understand why some players keep swinging for the fences while others walk off the field for good.
Drawing on seven years of community cricket data, DSI researchers developed clustering models to describe the behavioural patterns of community players and to identify the drivers that keep both kids and adults playing.
“For community players, there are many factors related to their intentions to return to cricket. For example, their age, where they’re living, their performance — all these affect their intention to stay in the sport,” says Dr Bin Liang, the DSI project lead.
Drawing on these factors, the DSI team developed machine learning models that consider the unique circumstances and history of each player to forecast their likelihood of disengaging from cricket. Around 70 per cent of the time, these models can now predict whether a player will hang up the gloves.
Via a series of user-friendly dashboards, Cricket NSW staff can explore engagement trends and forecast retention levels for the state as a whole, as well as for particular regions or specific players.
“DSI delivered a suite of analysis tools which empowered our team to make targeted and meaningful plans to improve the experience of all those who play cricket,” says Scott Peterson, Principal Data Scientist at Cricket NSW.
"The codebase was contemporary, open source and well documented, allowing a seamless deployment onto our Ludis Analytics platform. This enables our cricket managers downstream to access the DSI insights in a single click, anywhere in NSW."
These evidence-based engagement strategies are set to bowl the cricket world over — and they might just keep the next Don Bradman connected to one of Australia’s most popular sports.
