I am no expert, but I am skeptical because the paper results looks way too perfect to be true.

Then instead of doing statistical analysis with things like p-values or probabilities, they make wildly out of scope assertions like this:

> "Considering all pandemics with obtained extremum points, we can conclude that sunspot extrema coincide with the pandemics’ first appearance probably because of mutation on virus DNA or generation of a new virus."

They are claiming both sunspot minima and maxima, not just one or the other (which seems more intuitive?) somehow cause pandemics due to virus DNA mutations, even for pandemics that were caused by bacteria.

I am all for using statistics and data science for discovery of new phenomena, but I suspect this one is total bunk.

Correlations are to be found everywhere: http://www.tylervigen.com/spurious/random

Not to say there isn't a causal relationship, but a correlation graph is only a "maybe there's a relationship" and not "this means that".

There are so many visible red flags in this one and my priors are sturdy enough that I think it's safe to call it nonsense with barely a skim. This seems like somebody that came to a spurious conclusion and then decided to write a paper justifying it by spamming statistical techniques until something fit.

Can we get a vote-to-ban button for posters like this? It’s getting bad lately and clearly this has become too big a job for Dang.

That coincidence plot in the paper is amazing. I wonder what the criteria was to include a pandemic episode.

In the last link I don’t see any correlation at all, these are are completely different time scales? And even if there was, it’s meaningless as you can probably find a dozen random graphs (price of milk, number of eggs laid per chicken, left-handed births or whatever) that happen to be a near-perfect match to solar cycles.