When does Ad Hoc reasoning become a fallacy in regards to Physics?

I reference this to theoretical physics especially String Theory.

I won't go into the nitty gritty but at the heart of String theory is some serious AD Hoc Arguments that build on top of more ad hoc arguments. A few examples : In theories of particle physics based on string theory, the characteristic length scale of strings is assumed to be on the order of the Planck length, or 10 to the negative 35 meters, the scale at which the effects of quantum gravity are believed to become significant. If any of you are wondering to get down to Planck lengths, you need a particle collider that is  a quadrillion times stronger than the maximum LHC energy. Its essentially unfalsifiable as we might never be able to even remotely reach those levels and even if we do theirs ways to explain away how come we can't see it.

String theories require extra dimensions either higher or compactified and another object called Branes. Essentially in certain models our universe exists inside a higher dimensional hyperspace that operates like a membrane or these particles are compactified on such small scales in such a way that they are existing in higher dimensions. Gravity being what comes through the membrane in brane theory models . The problem is that the math on these things are not solid and and we haven't see them and similar to strings you can always come up with something to explain why you haven't see them yet. 

This is all very under simplified I know but a similar thing is called Preons. It was a alternative model with a different particle to the Higgs Boson. Even a decade in with the discovery of the Higgs these models are still being researched on despite making predictions that turned out to be wrong and having Evidence against the model but then try to explain away and keep the model from being wrong. 

There are scientific theories that are so well supported they are often synonymous with "fact" (e.g. the theory of evolution vs the fact of evolution). Then, there are theories, like string theory, that are very speculative. A key to rational thinking is to not only hold all theories provisionally , but only to the degree supported by the evidence. Physics is no different than any other topic in regards to when ad hoc reasoning is involved. Remember that ad hoc reasoning is not a fallacy per se, but can be when used as an argument (or as a substitution for a valid argument). The critical thinker can use ad hoc reasoning when hypothesizing or building a theory, as long as it is recognized as such. It is also important to recognize that the more ad hoc reasoning is needed, the weaker the theory.

When does Ad Hoc reasoning become a fallacy? Technically, when used to replace an argument. For example:

Person 1: String theory has many problems and is likely not an accurate representation of how the universe works.

Person 2: But it is possible that (addressing possible solutions with no evidence) so therefore, string theory is likely true.

The main problem comes back to believing in a proposition without enough evidence to warrant the belief (or believing it to the degree that is unwarranted by the available evidence, or lack thereof).