Bayesian statistics require priors

You need to have some kind of Prior, which is your expectation.

They can be strong or weak (in other words Uninformative priors)