/snip
So, here's my argument in a nutshell.
If Player A came to the table with a die rolled character with 6 12's, most DM's would let that player reroll that. Even though that is 100% a baseline character, there are a significant number of DM's who would consider that "unplayable" or at least really underpowered.
If Player B came to the table with 14, 14, 14, 12, 12, 12, most Dm's wouldn't even bat an eye. That's perfectly fine and no one would question that, even though that's actually a 76 stat character and significantly more powerful than a baseline character.
If Player C came to the table with 12, 12, 12, 10, 10, 10, that's pretty much a guaranteed reroll. It's a pretty rare DM that would insist that those rolls stand. Even with 4x12 and 2x10, even though that's a 68 stat character, would likely get rerolled.
So the floor for die rolled characters is probably about 70. There might be this or that character that's lower than that, but it's pretty rare. And, an entire group of 70? Never going to happen. Is there anyone out there that can honestly say that the highest stats in their group was 70? Ever? I'm certainly not going to hold my breath on that one.
For all the arguments about "exploring characters" and "organic play" and whatever, it seems that all that goes straight out the window when the rubber meets the road. Generous die generation methods mean that most characters will be above baseline, and probably significantly above baseline. Groups will always average above baseline.
When the floor of die rolled characters is effectively baseline or higher, I don't find it particularly difficult to come to the conclusion that die roll generation is based around having higher than baseline characters. Which is perfectly fine. Play what you want. But arguments that it's all about "exploring the character" and "wanting weaknesses" and "weaknesses define the character" kind of fall flat when the majority of die rolled characters actually don't have weaknesses and are in fact stronger than baseline.
And, finally, again, for evidence, I direct the reader to their own die rolled groups. Simply add up the numbers. Is your group below baseline on average? How about your last group? Or the one before that? Or the one before that? Have you ever played in a group that averaged below baseline?
@DND_Reborn said that only 2% of groups would average below baseline using 4d6-1. Are you part of that 2%? And, doesn't that mean that 98% would be baseline or higher?