The so-called Poisson probability distribution allows you to calculate the likelihood of an event happening. It’s necessary at the start, however, to know roughly how rare the event is. But if you do know that, as far as I can gather from John Allen Paulos, Innumeracy: Mathematical Literacy and Its Consequences (Hill and Wang, 1988) at 48, you can use this frequency information along with the Poisson formula to get a quite accurate idea of the event’s occurrence.
For example, in what percentage of the years to come there will be a class-action lawsuit, a disabling conflict of interest with your primary law firm, or an acquisition larger than $1 billion. If a number of companies pooled their data, there would be nothing fishy about a Poisson calculation.