The Fansure Index

Introduction:

How to predict the quality of upcoming games is a recurring question asked by fans, TV networks, ticket vendors, the League Office, and all stakeholders alike.  Fans want the quality of their experience to match the ticket price they pay, while ticket vendors price games based on the perceived quality and demand.  However, unforeseen circumstances such as star player unavailability can adversely impact these dynamics, especially for fans paying top dollar with the primary intent of seeing a star player in action.

Using machine learning, Fansure has developed an index - the Fansure Index - to help fans be more informed as to the expected quality of the game they are interested in attending based on 1) performance parameters for teams and players (on- and off-court) and 2) the likelihood of star players being available for the game.  In other words, we can predict the quality of games based on both known and probabilistic parameters.  The index can also be used by ticket retailers and TV networks when deciding on ticket prices and which games to air, respectively.  Furthermore, it can be used by players to assess their potential market value for a franchise.

Methods:

The Fansure Index – developed by two NASA rocket scientists – uses team and player rating systems to quantify the aggregate star power and market appeal of every NBA team. The combined ratings of two teams in any given matchup are used to generate the Index on a 1-10 scale. Simulations show that ticket prices in the various franchise markets correlate exponentially to the Index to a high degree of statistical significance. This tool can be used to predict the ticket price impact for various scenarios. If star players switch teams, the Index can predict the corresponding change in ticket prices (e.g., Kawhi Leonard and Jimmy Butler joining the Clippers would increase their prices by 229%). Similarly, if a star player gets injured or traded, the ticket value lost for subsequent games can be quantified.  The probabilities of key players missing games is modeled using a logistical regression algorithm. Factors considered in this model include schedule-based parameters and player-specific metrics. Knowing the ticket price impact of available key players on a game can greatly help ticket vendors make market pricing decisions accordingly.

Results:

The figures below show nominal ticket prices for select teams based on the Fansure Index (left) and the predicted probability of the top NBA players missing games over the past five seasons (right).             

 Plot of Ticket Prices vs. the Fansure Index for some of the league’s top teams

Plot of Ticket Prices vs. the Fansure Index for some of the league’s top teams

 Plot of average probability of games missed for the top 30 players over the course of the last several NBA seasons.

Plot of average probability of games missed for the top 30 players over the course of the last several NBA seasons.

Conclusion:

The Fansure Index is a metric for assessing the quality of a game using team and player attributes and predicted player availability. Using ticket prices for the 2018-19 season, the Index accurately conveys the value of games. This tool can be used by fans when prioritizing which games to watch or attend and by those with business involvement such as ticket vendors, TV networks, and players.