I mostly agree, but I think you are confounding two things here. The actual argument doesn't have to contain a better alternative than the current best explanation. It's enough to show the errors of the current explanation - as long as you are doing it using the scientific method.

I.e. it's enough to point out flaws in a scientific theory even if you don't have a better one yet. E.g. if talking about covid, it's indeed valuable to prove that the virus does not use the receptor, suggested by the status quo, to get into the cells. One could do this without identifying the actual receptor it does use (by showing that it can infect cells that lack the said receptors). Another example would be showing that a specific medication that is thought to work, actually does not. (Say hydroxychloroquine...)

> You can't just say, "Science has gotten it wrong in the past so it probably has got it wrong now, and therefore you should pay attention to my crackpot theory."

Now I fully agree with this. But where these guys go completely off the rail is that they try to discredit the scientific method by saying that it's wrong, because it provided erronous results in the past. And try to use this as an argument for why their suggested alternative solution doesn't need to pass the test of science.

And here is where they can be challenged with having to come up with a proven better alternative to the scientific method. (Which is kind of nonsense, because science is a self-improving/self-correcting system anyway.)