The word "evidence" is badly misunderstood, often even by the very people throwing it around.
There's a rather large contingent of people who use it as a synonym for "proof", thus demonstrating their complete lack of understanding of the very concept even as they think they are looking smart. There's a lot of people who think that it is somehow the responsibility of someone like Carlson to have unambiguous proof that is often physically, legally, or morally impossible for them to have. Not that the Kantian imperative is the be-all, end-all of morality or philosophy, but it's still a useful idea, and if one runs this memeset through that grid it quickly reveals that you can't use this as the bar people must meet. It creates a society where anything that can be hidden, often not even hidden particularly well but just hidden at all since the standard is "proof", will be gotten away with. This is impractical.
(Although there is an extent to which we do indeed live in that society.)
There's another smaller contingent (from what I can see, the former is the dominant group) who thinks that "evidence" doesn't have to be proof per se, but that for something to be "evidence" it must be single-handedly capable of meeting some bar, be it "preponderance of evidence" or whatever is being used. These people are blind to the fact that it is often possible to produce ten pieces of evidence that are individually weak, but together are too improbable to ignore. While this case must certainly be treated with care since it is sensitive to how well done the Bayesian-style analysis of their probabilities is, it is perfectly reasonable to use such evidence to come to conclusions. This contingent will often try to defeat the evidence in detail, decrying each individual one and downplaying it, and completely failing to address the totality, be it either as a method of being disingenuous, or, I am convinced, often just lacking the intellectual firepower necessary to address the totality of evidence holistically.
It's important not to fall into that trap because I think in the real world, this is how many very important questions can be settled... and I don't just mean big philosophical or societal questions, or even news of the day like this, but even little personal things. The signs that your partner is cheating on you often come in like this... you may not catch them in the act, nor even receive some individually huge clue, because there's one or two humans intelligently trying to deceive you. Such situations often end up giving off a lot of little clues, each of them individually explainable-away, but the totality pointing to the truth. Or you can determine some new direction your company may be trying to take by a pattern of little changes long before they announce it publicly. Any number of things in the world where you can carefully assemble a lot of little things to come to some reasonably strong conclusion.
Back on topic, the combination of the fact that Neimann was known to have cheated before and the accusation of the current best player in chess, who was kind of going out on a limb to do so, was quite strong evidence that Niemann did cheat. Neither individually would be enough for me to come to that conclusion strongly on its own, but the combination was compelling. I considered that enough proof for my personal opinion. Proof enough for authority figures to sanction would require a higher standard and I support and applaud them for being more careful.