Private, I respect your opinion and research on this, but the guy does say R6 is better in low light. So I am confused, Again!
Ok there are several approaches you can take to answer this question satisfactorily for yourself, but first you have to quantify ‘better’. For me it means a viewable improvement in IQ at standard screen or output sizes, for Canon it might well mean measurable on an oscilloscope. If you agree with Canon then technically they might be right, if you agree with my definition then that small a difference, ‘better’, is irrelevant.
Then you can look at the numbers on Photonstophotos.com, compare any same generation sensors from any manufacturer in the last ten years that illustrate a meaningful correlation to fewer pixels equates better high iso performance, I haven’t managed to find a set. Personally I love the site and find my actual image performance aligns very closely with their numbers and comparisons, but it is a bit scientific for a visual medium and detached from what we are actually doing.
For a true visual reference DPReview has a huge library of downloadable RAW files. Just download a few you are interested in, or that can prove or disprove an idea you have. I have downloaded loads of RAW files from them exactly for instances like this. So once you have your files you have to define a realistic comparison, for me that is same sized output, it is after all pointless comparing different pixel density sensors at a percentage value, they have to be same size, or ‘print’ value as that is what we are actually interested in seeing. If you want to print to 20” x 30” then upsize the smaller resolution etc.
You could also listen to and believe anything you are told, but if you do that consider the source, and where they got the information from. The camera industry is like every other industry, it is trying to sell product. In their case it is a tech product that few people understand the physics of and unfortunate marketing or memes that were normally started to help alleviate confusion really end up causing more confusion down the line when they become accepted norms. Things like crop camera focal lengths, we know the focal length doesn’t change but for an uneducated market it is easier to name focal lengths incorrectly rather than explain the definition of field of view. Telephoto effect/compression is another, there is no such thing but it is easier to introduce that meme and then teach it rather than explain perspective.
Now back in the day before on sensor electronics were refined to the efficient and size level they have been, and back sensor illumination was a thing, and gapless micro lens technology that focused nearly 100% of the light hitting the sensor area down into the actual photosensors regardless of the electronics also on the face of that sensor there was some truth to the idea that fewer pixels on the same area had less noise. But those improvements have been made and the light gathering efficiency of the sensors regardless of the pixel density is so close to optimal there is practically no visual difference in noise in same sized output.
Certainly I haven’t seen a difference big enough to get different pixel density cameras to cover different uses and I personally compared the 1DS MkIII and the 7D a million years ago! Since that era I haven’t seen pixel density/noise differences that can’t be post processed away, and despite my many postings illustrating that point with copious links and actual images and downloadable files I have never had anybody refute it with actual images. And I am certain that will be the case with the R5 and R6.