LucidFlight wrote: However, given a posterior distribution P(θ|y) for quantity z of fried chicken thought to have been consumed, there is a estimated 100% chance of KFC resulting in a number 2.
Did you factor the strength of the posterior tensor into your flow calculations?
I guess Lucid did not. The whole KFC angle looks fishy.