The number is the RECIPROCAL of the infection fatality rate of H5N1: if the market resolves to 300, it means that for every 300 people infected, 1 dies.

Assuming there is an H5N1 pandemic that infects at least 10,000 people worldwide, what will Manifold's best estimate of the infection fatality rate be, one year after the pandemic starts?

(Will resolve to IFR based on conditions in rich Western countries at the start of the hypothetical pandemic. This means IFR among unvaccinated people, unless somehow people get vaccinated before the pandemic starts.)