Business Performance Simulator Accurately Predicts Season Outcome for Major Sports Franchise
Overview and Problem Summary
In April of 2020, the pandemic was raging across the United States, causing businesses closures and massive uncertainty. The seasons of all the major sports leagues had been put on hold. A major league sports franchise in the northeast was struggling to forecast their financial future in all of the uncertainty. They needed to have some basis upon which to forecast whether the season would be played, how many games would there likely be, would fans be allowed, if so how many would attend.
The model has configurable variables for key business metrics such as:
- Venue capacity
- Average revenue per attendee
- Player salaries
- Fixed venue and team expenses
- League and television revenue
In addition, BPS tracks the state of government interventions such as capacity restrictions at the city level. This data is fed into a proprietary model of pandemic spread to enable predictions on the lifting of or imposition of restrictions.
The Business Performance Simulator (BPS is ideal solution for predicting the performance of any business during times of uncertainty. Enabled Concept utilized BPS to model the key attributes of the sports franchise and the simulate how it was likely to perform from May through September of 2020.
Using those drivers and metrics, the BPS uses data science to generate a collection of possible business conditions and simulates season under those conditions, producing three key forecasts:
- Pandemic end-dates: Using cutting edge data science, a forecast is generated of possible dates the pandemic will end and the shutdown lifted for in each of your locations.
- Season schedule: If a season can be configured, the home and away games are scheduled and fan attendance is forecasted for each game use an attendance recovery curve.
- Financial performance: Using each of the possible end-dates, a monthly P/L is created, each forecast allowing for some variation in your key metrics.
BPS simulated the season thousands of times and produced a distribution of probable outcomes. The 90% confidence interval correctly predicted that the season would start in the beginning of July and would proceed on an abbreviated schedule with no fans. This scenario was forecasted to produce millions of dollars in losses for the team. Armed with this information, the team was able to renegotiate key expense contracts and acquire sufficient liquidity to weather the storm.