It's bayesian reasoning. If I tell you that I predict the sun will explode tomorrow, you should consider the sun's track record of not exploding when evaluating my claim.
Also, using only past billion occurrences of the sun not blowing up, and then still concluding that the sun will not blow up despite of any recent indications of the sun showing anomalous activity seems a more accurate analogy.
> despite of any recent indications of the sun showing anomalous activity
The whole point is that the videos are not in fact the indications of anomalous activity that they're made out to be. There are mundane explanations for all of them.