Seriously? Again?
Everyone rolls the same dice, using the same method, that's fair.
The results may be radically imbalanced PCs, but the method is perfectly fair.
ASIs do even out an imbalance in primary stat over time - they shift it to secondary or tertiary stats, which are lower impact, so it's not erased, but it's reduced in significance.
View attachment 83177
Array is also clearly a player option in the standard rules. So you can't guarantee that someone won't be OP, but you can at least assure that you're not sub-par.
Everyone rolls the same dice
just the once, and have to live with the results
for the rest of the game.
I will answer each of your points in turn.
You make the point yourself, oddly. I suppose that it is because it is beyond obvious that it is the results of the method we are talking about here, not the equity of how many dice you roll. Stop deploying smoke and mirrors to defend the inherent unfairness in random stat generation. Maths is maths, no matter how much you might like stat rolling.
ASI's do not even out if Feats are used in the game - which I have already stated, but you didn't want to answer... Also, don't try and pretend Hit Points, Skills and Saves don't matter, which make any secondary stat allocation of great use to any class,
especially in a Bounded Accuracy ruleset.
Your attachment makes my point very well - thanks for including it.
Only if your DM allows you to substitute the standard array AFTER failing to get a decent set of random stats.
I might as well say, "You are defending random stats, seriously!! AGAIN!!??" in answer to your first line. The game already has a great deal of randomness. But if you were to put it another way at the game table to people they wouldn't be happy.
If I were to have everyone with a point buy character then roll a 1d4 and minus 1 from the result, and have to give up the resulting number of stat boosts at the appropriate levels, most people would think that was unfair, and stupid. Similarly, you could allow non-fighters to use the Fighter stat boost advancement track if they rolled a 1d6 and got a 6. Or you could just have each person roll and give the highest roller an extra 5 points to spend on stats.
There are many ways to do random stats and simulate the consequences for a game. But you wouldn't get anyone to agree to those - they aren't 'tradition'.
If tradition and the fun of the initial randomness is more important than fairness, then the rules allow you to go that way.
I don't like forcing disadvantage on people, or allowing some to have baked-in edge over others for the full course of a game. Everyone should be equal at the start. It's CHOICE that should make a character, not the luck of the dice the first time you pick them up.