One can say: there are two possibilities concerning weather: either sunny or rainy. Therefore, the likelihood of a sunny day is 1/2.
But we know that it is FALSE: the likelihood of a sunny days is de facto higher in a sunny country. We calculate the likelihood of a sunny day not a priori, but a posteriori, by collecting data. And the likelihood we find expresses the laws followed by nature.
The defenders of the fine-tuning arguments are wrong because they calculate the likelihood of some cosmic phenomena in an a priori way, but they should instead collect data and check whether there are natural laws which make some events likely.
They could object: we concede that the a posteriori way of calculating the likelihood of phenomena is better than the a priori way, but since we can’t use the a posteriori way, we can only rely on the worse method and determine what is reasonable this way.