# weighted bayes theorem

I am using a simple implementation of Bayes theorem to find the discrete probability distribution of proportion of wins.

naiveBayes <- function(theta,prior,win=T)
{
ifelse(win==T,likelihood <- theta,likelihood <- 1-theta)
constant <- sum(prior * likelihood)
posterior <- (prior * likelihood) / constant
posterior
}


A win being True or FALSE gives a posterior which I will use a prior for the second computation and so on.

Question: How to give more weight to recent event? Say, the new event is a loss (win=F), I would like it to have higher influence than what the bayes theorem provides.