Performing Maximum A Posteriori estimation on a set of dice results.

Question

I have a set of data obtained from rolling a 20 sided dice 1000 times. I understand that ideally a dice would have a uniform distribution, and that forms my prior belief. But how exactly does one go about finding the MAP estimation that maximises the likelihood of me observing this set of data taking into account the data itself and the prior belief? Many sources have varying approaches like converting each side's result into a binomial distribution and using beta distribution as the prior, which is completely foreign to me, and the wiki says the prior to uniform distribution is the pareto distribution, while some other sources says its a flat prior which is very vaguely explained. Could anyone shed some light on how I could approach this? Preferably with as much detail as possible.