AI and Ticket Competition

1. Choose your role

Choose a role to begin.

2. Market setting

Official ticket price $100

Expected resale price $180

Fans in the market 200

Scalpers in the market 50

3. Predict behavior before seeing the model result

In an effort-based ticket contest, how much effort would you choose? If AI lowers your participation cost, would you increase your effort? Which mechanism do you think better protects genuine fan access? Why?

4. What does our model predict?

Effort Contest

In the classical Tullock benchmark, lower effort cost increases equilibrium effort.

For a prize value of 80 and 10 symmetric players:

Cost = 10 → equilibrium effort = 0.72
Cost = 5 → equilibrium effort = 1.44
Cost = 1 → equilibrium effort = 7.20

In our synthetic asymmetric-cost model, AI can shift effort strongly toward scalpers when their participation cost falls much more than fan cost.

Verified Lottery

Without AI, scalpers use about 1 identity on average.

With lower identity-management cost, they use about 4 identities on average.

In our simulation, primary fan allocation falls from about 80.6% without AI to about 49.8% with AI.

Each simulation represents competition for one representative scarce ticket. These results come from a synthetic model with stipulated parameters. They are not empirical estimates of real ticket markets.

5. Reflection

Did the model result match your prediction? Which part surprised you most? After seeing the result, which mechanism would you choose as a platform designer?

Project connection

Our project studies how AI changes the cost of strategic participation in ticket markets.

We compare two mechanisms: an effort-based contest and a verified lottery.

Our central question is whether AI creates an asymmetric cost advantage for professional scalpers, and whether mechanism design can reduce that advantage.

Primary allocation and final ownership are reported separately because resale can return tickets to fans even when scalpers win them initially.

GitHub exact commit: 91cd8d1