The answer you cited presumes, wrongly, that it is possible to assign probabilities to theories. A probability has to be calculated. There has to be an explanation of why some particular assignment is made and not another assignment. That explanation can't be assigned a probability because it is a precondition for calculating the probability.
There is another way of putting this problem. A probability can only be assigned to elements in a space of events. A theory is not an event. General relativity, for example, is not an event. An event is something that takes place in a particular region of space and time. General relativity is supposed to apply to all events everywhere in space and time and it is an account of what is happening in reality to bring about those events.
There are a lot of problems with trying to assign probabilities to theories. For example, if you come up with a bunch of arguments pro and con for some set of theories and you try to attach weights to the arguments and use those weights to apportion credences to theories then you end up with problems similar to the apportionment paradoxes that are well known in the literature on voting systems, see Chapter 13 of "The Beginning of Infinity" by David Deutsch.
There is a more fundamental problem. Let's say you have a theory and you come across a criticism of it in the form of an experimental observation or whatever. Either the theory is wrong or there is something wrong with the criticism. There is no third option. So you should be trying to work out which of those two options is correct. Bayes' theorem can be used in the context of such an investigation if the relevant theory makes probabilistic predictions. For example, a person saying that an experimental observation is unlikely according to some theory might be wrong because he didn't use Bayes' theorem to correctly predict conditional probabilities from the experimental setup, but that's not the same as assigning a probability or credence to a theory.
There is a further problem. Probability doesn't describe uncertainty. Uncertainty means you don't know if X is true so how does that imply assigning a specific number to X as a probability. For example, at present the red shift of distant galaxies doesn't conform to what we would expect if the vanilla big bang model was true. There are various proposals for what is happening in reality to produce those results that haven't been ruled out. We just don't know which of those proposals, if any, is correct.
In reality, our theories about how the world works are guesses and any one of them might be wrong and there is no way to know about it being wrong until we find a replacement. And observations are in a similar situation because any experiment involves an explanation of how the experiment works and that explanation might be wrong. In the light of this Karl Popper pointed out that all of our knowledge is guesswork, see the reading list here
https://fallibleideas.com/books#popper
The way we assess our guesses is by looking for criticisms of them by looking for independent implications of those guesses, e.g. if you think your telescope has a bad lens and this explains a wrong prediction you made about what you would see when you looked through the telescope, then you might point the telescope somewhere else to find out whether it is distorting observations made by other telescopes or you might replace the lens.
As Elliot Temple theories should be given a yes or no status
https://criticalfallibilism.com/yes-or-no-philosophy/
https://criticalfallibilism.com/yes-or-no-philosophy-summary/
You should try to come up with criticisms that decisively rule out an option instead of weighting the options. Trying to assign probabilities to theories isn't a good idea because if you're going to decide between options X and Y saying something vague about weighing and probabilities, as Bayesian epistemologists do, doesn't distinguish one of the two options and so doesn't explain your decision.
