Have you looked at the summary table in the linked sheet?
Yes... After all, I am the one who put the time and work in to record, chart and make it.
To what problem are you referring?
To the fact that, even though, in the long run, the end results deliver the expected averages, overall, the RNG is deeply flawed in its production of its pseudo-random numbers - and creates a visible wave pattern strong enough that you can actively see its algorithm working; I don't wish to believe that anyone posting here is simple-minded enough that they cannot see why this is a deeply damaging problem for the RNG of a game whose every mechanic is based around using that RNG... You aren't that stupid, I know you're not; no-one posting here is.
What problem? the BG3 chart looks completely fine with this 200 roll example. It almost hits the perfect average. Each group of 5 (1-5, 6-10, 11-15, and 16-20) are all near 50 rolls. This comparison doesn't highlight any problems with BG3s RNG.
The poorly modulated algorithm that shows, very clearly, the wave pattern of it working without reseed or refreshing its chaos intakes (presuming it has any; it doesn't look like it does). You CAN see it, can't you? When you LOOK at the charts? The uniform wave that continues reliably in a fixed pattern that consequently ensures then when you are in the low swing of the algorithm, you are likely to have successive low rolls, and when you are in the high swing, you are likely to have successive high rolls. This means, by natural extension, that you are MORE likely to miss with advantage when you are in the low part of the algorithm, overall, than you are with a more naturalised RNG, and you are more likely to have functionally 'wasted' advantage as well, when you're in the high end. Creatures attacking you with disadvantage while you're in the high end of the algorithm are more likely to hit you anyway due to successive high rolls, than they would be in a more naturalised RNG; the problems swing both ways and cut in both directions, naturally, but they do cut, and it's a very real, very tangible result of having a weak rng that displays a clear wave pattern.
With this RNG you will see far more instances of "failure against all odds" and more instances of "repeated success with tiny odds"... which is part of the result that has led to the sheer volume of threads about people complaining about the dice on this forum.
This is not confirmation bias; this is a statistical truth that is the result of having such a poor RNG compared to other similar games. This is the root of their dice problem; this is what needs to be fixed, before any other solutions are hacked in as band-aids.