UW's @trvrb estimates ~570 cases in WA, a state of 7.5 million. So an event with >9120 people chosen at random from WA has >50% chance of having at least one case.
That's how super-spreading happens. Here's the math.
As the proportion of infected gets larger, even a relatively small "event" like a corporate office with 100 people has significant risk.
Reason: while molecular evidence gives us a direct estimate of ~570 cases in one cluster in WA, we're seeing community spread now that testing has finally begun.