What can some simple Bayesian analysis tell us about vaccine safety

Steve Kirsch posed the following problem: He has a dataset of 7500 people of whom 5625 were vaxxed; 1875 were unvaxxed Over a period of 6 months (182 days) 15 people died, ALL from the vaxxed group Of the 15 who died, 4 died within 24 hours of getting the vaccine. Steve asked what’s the probability of this happening if the vaxx was safe? (update: Steve has now written up about the context/source fot this dataset: ) Here's how I answered it using Bayes Theorem to combine the different pieces of evidence. (Note: this assumes the data is accurate, ignores all confounders like age, and assumes the vaxxed and unvaxxed are similar types of people) Software used for the Bayesian network analysis is