I read them with some caveats in mind:
1. I did not see much change at all when I clicked 265/70/17, 265/75/16, 235/70/16. So I don't think there is much change just by size.
However, remember that size can infer the type of vehicle and year of vehicle and various capabilities of the vehicle. 265/70/17 is the normal size for current model Expiditions/Tahoes, CHevy, Ford and Dodge pick ups. Those normaly have 4x4 and normally be used for work and off road. Size like 235/70/16 went on base model CHvy S-10, base model Rangers, small pick ups and small SUV like the Blazer--not the sort of vehicle owners will take to the bakc wood or hunting or in bad weather. SO owner use will affect whether the tire is exposed to rough usage. Sizes like 265/75/16 is a 1990s size and back then they had ABS, but normally not traction control VSC, uphill/downhill control, etc. Just that difference can make a tire appear to be better or worse but it really is the car's mechanical and computer technology and not just the tires.
18-20 inch tires will no be normaly used in 1980-1990s vehicles. They are current models--with all tjhe bells and whistles to enhance traction. And more likely to be found on Infinity F35/45, Lexus, Land Crusiers or Range Rovers...hardly the type that gets dirty. Again ,affecting how owners perceive tire performance.
2. There is probably as many different types of snow as Eskimop words for it. Slush, hard pack, Sierra cement, deep dry fluff, heavy wet chunks. Melted nad rechilled stuff. So when a tire is rated for snow use..we really don't know what type and how it is used.
3. There are just as many types of mud. Water and dirt, water and gravel, dirt is nto the same as clay, and clay may be just wert enough to be like playdoh, or Lousisiana gumbo, or Georgia brown cement--slick, greasy, and hardens to concrete. So we don't know what surveys means when someone complains abotu mud performance.
4. Most importanly, how the owner drives the vehicle. that is the biggest most unpredicatable unknown. A good careful experinced 4 wheeler can make a two wheeler do better than a newbie on a 4x4. AWD may operate better than part time in certain conditions. A lead foot may not do as well as a seasoned driver.
Again, that affect the perception of how a tire performs.
5. The honeymoon. A new tire will work better than one with 50k miles on it. A new tire like the Pirelli ATR will almost alwyas be rated within 25k miles, while an established brand will have many milions of miles of experience behind it and drivesr commenting on it after running it down to the treads.
6. We see more dry traftion stories now since it's spring, summer. We have fewer winter stories to evaluate. IN deep winter, we'll see more of that and less of warm weatehr performance.
So...........with so many variables one could and would expect that anecdotal evidence, not even empirical evidence as there is no controlled conditions, would "wash" and some level of consistentcy will emerge.
I would look at flaming stories with a critical eye, exceptioons are to be expected, but at the over generall PATTERN. Some pattres nwill coelace and start sharpening their foucs on good and not so good points.