Whether the proposed mechanism for a therapy makes any sense given what we already know about the world - from common sense, and from the sciences should prejudice our judgment about the chance an alternative therapy could actually be effective. I am not referring here to specific evidence for the therapy, but rather whether it has a reasonable rationale - whether it is plausible that it could work.
As an example, if you are told that an alternative therapy works because it stimulates the immune system, at a very superficial level, this is a possible rationale - immunotherapies are known to work against some cancers (whether there is evidence to support the rationale is another matter). But if you are told that the therapy works by shooting a ray of a form of "energy" which is otherwise completely unknown to science, you should be very very skeptical of the therapy, and you should require far more evidence to judge it as promising than you would something which has a reasonable rationale. Rather than ruling therapies in or out directly, the plausibility raises and lowers the estimated promise of the therapy given the same strength of actual evidence for the therapy. Plausiblity can be extremely difficult for the layman to judge because many alternative therapies have a very technical and scientific sounding proposed mechanism of action which make no scientific sense whatever. At the same time, the universe is full of surprises and unexpected discoveries. One academic paper I read rated the chance that the Gerson dietary therapy could work as zero because it was considered to be completely impossible that a diet could affect cancer. So no matter how strong the actual evidence the author would not believe the Gerson therapy could work. I find this vastly unreasonable. Even if there is no diet known to cure cancer, diet does affect the structure and function of the body, and a large body of evidence suggests nutrition plays a role in the development. I think that one should keep relatively open mind about what is possible, but far from an empty mind. When there appears to be no reasonable or rational explanation of how a therapy works, that should be taken into account, and where the therapy fits in with other things which are proven that should also be taken into account. The more radical the explanation is, the more it relies on deep and previously unknown scientific principles, the more evidence for it we should require.
An example of a therapy I find implausible is homeopathy. In homeopathy a substance is diluted in water, 10 or 100 to 1, and then shaken. Then it is diluted again and shaken again. This process is repeated many times. So many times, that as proponents of homeopathy will cheerfully admit, it is mathematically a virtual certainty that not even one molecule of the original substance is present in the final product. So the obvious question is whether it is plausible that this process results in anything that is really different from ordinary water. Certainly this does not agree with anything we experience in everyday life - it seems to violate commonsense. The proponents of homeopathy may talk vaguely of some sort of "energy", "memory", or "resonance" remaining in the water, but there is nothing in mainstream science to account for such a thing, and there is no strong evidence that this energy can be measured independently. Homeopathy requires a radical explanation at deep levels! Because I find homeopathy implausible, I would require much stronger evidence to judge it as promising than I would a treatment which is more plausible.
An example of a therapy I find relatively plausible is anti-neoplastons. Anti-neoplastons are well defined chemical entities, of a family which are known (through credible research) to have biological effects against cancer cells. I find it very reasonable that a drug-like substance could have an effect on cancer. Deciding whether it does or not requires looking at the evidence.