Most old video games were a normal difficulty progression (either by levels or over the same amount of time) independent of player skill, which is why some players would die on level 1 and others could get to level 100. Adaptive difficulty games would be more like a game that if you beat level 1 quickly with perfect accuracy, the enemies in level 2 would suddenly be a lot tougher and more accurate, while if you barely completed level 1, the enemies would only be slightly better (or possibly even easier) than the enemies in level 1. So the end result would be something like the good player and the bad player both probably dying around level 5, and both able to complete the game after dumping in around 10 quarters for continues. As a result, there is no bragging rights for making it to level 5 since everyone makes it to level 5, and the only bragging right for beating the game is that you were willing to dump in enough quarters.
In the games I've been in, the party usually wouldn't know if it was a 20th level Medium encounter or a 5th level Medium encounter, they just know how hard it was to succeed, how long it took them, and what it cost them. Even if they did know, it's not like there is anyone they can brag to that would care what level encounter they defeated. If they come up with a new tactic that they think will save them time and/or resources, and suddenly the encounters get correspondingly harder so that they are still spending the same time and resources while using the new tactic, what incentive would they have to put the effort into continuing to improve? The GM is deciding beforehand the whether or not the the PCs' abilities will be effective in a given encounter and how difficult the encounter will be, so it is the GM's decisions that matter to the success of the encounter, not the players' decisions. At some point, the GM might as well just hand-wave the encounter, narrate it to the players, and tell them how many resources to mark off their character sheet. That would be a much faster route to the end result of the encounter tailored according to the specific party's abilities.
Not true at all. No matter how much you tailor encounters, random dice rolls and player choices always take things in unexpected directions good or bad.
If you carefully design an encounter to be only somewhat challenging, bad dice rolls could make it deadly. You could design an encounter to be deadly and good dice rolls make it easy. Or one unlucky missed save taking out a bad guy. Or one unlucky save by a PC making the fight a ton harder. Tailoring only gets you so far. Random dice rolls and player tactical choices can take the encounter in unexpected directions and almost always do.
The fun isn't always winning or losing combat. It's figuring out why you were in the battle to begin with and watching a story unfold. At the moment one of the characters I'm running hates orcs and has killed an enormous number of them. He is known amongst the orc tribes as an orc killer. Every time he fights orcs, they are trying to kill him. Right now he is on a boat with half-orcs that know who he is. I'm playing it up big time that they hate him and fear him. He's playing it up right back doing things to scare them. It makes for great role-playing regardless of how combat turns out. You know at some point I have to make some harsh orc nemesis for him that can actually challenge him, then let the dice decide the fate of both.
You seem to be confusing tailoring with automatic success or failure. That doesn't happen in a game with random dice rolls and player choices. Tailoring just means making the encounter interesting for the PCs and having it add to the PCs story.