I think you are focusing on changing the outcome of a combat; the strawman here is the thought that people who adapt encounters mid-flow are railroading GMs who want their version of the story to occur, no matter what. But that's a rarity. Most of the time we don't want to modify it to enforce an outcome (I certainly never have done so and based on comments, others rarely do), we do it to make the encounter itself more fun.
No, the only strawman here is you think that because I choose not to do it that I don't know how to do it, or you think that because I choose not to do it that I must have some different goal in mind than you do. Of course, it's not to change the outcome of the combat per se in the sense you define of "making the players lose". Even in the case of the Bounty Hunters versus Dr. Fist the mad robot scientist I brought up, my instinct to fudge was not about trying to get the players to lose, but to lengthen the chase out further. The reason either of us fudge is to lengthen or shorten the combat to make it "more fun". Or as you put it:
"to make the challenge closer to how we had envisaged it."
We? We? How many people were involved in planning this combat?
And oh boy does fudging to make an encounter closer to how you imagined it would be exactly describe the very process that I discourage novice DMs from fixating on.
And as for how to fudge, yes, I'm aware of pretty much all the techniques you describe.
As for why our two experiences are wildly different I'm going to assert entirely without evidence some theories that probably aren't fair, but not feel that guilty about it given just how grossly unfair your assumptions were about me:
1) I'm probably better at encounter design than you are. I don't need to do this very often in order to make exciting fights, possibly because I've been consciously trying to forgo the crutch of "fixing" the encounter on the fly for at least the last 30 years.
2) I play systems that are more lethal than what you play and I'm already pushing the edge of that envelope harder than you are. Those tweaking things upward 75% of the time would almost certainly end up with dead PC's quite often in my games.
3) I am playing more often with players that prioritize "Challenge" as core aspect of the aesthetics of play. If I play with players that don't do that, instead of fudging I change systems entirely to play a system that is less lethal, less fiddly, less gritty, etc. If I'm playing with kids that would be emotionally scarred by an injured animal companion, we won't be in a system where that can happen. But in groups that I typically play with, there fun comes from "beating the puzzle" against the (apparent) odds much as the hero in a book or movie seems to win against the odds. But while they want to just barely win as the most fun outcome, the players enjoyment of the play would be harmed if they realized that I was manipulating things so that they were steered to certain outcomes. One way a player can recognize that illusionism is if the GMs plans for the encounter never quite seem to go awry. The existence of encounters that obviously don't go as planned is sort of the proof in the pudding that the game isn't rigged, along with rituals like always rolling "important" rolls in the open (such as critical saving throws by either the PCs or the NPCs). The fact that the PC's get walkovers sometimes, or get their butts handed to them sometimes, or that encounters get grindy sometimes, or whatever proves that the times that the fight with the BBEG that goes down to one roll by one player when half the team is on the ground bleeding out are real and thus makes those victories have more savor in the way that if I had illusionism going on they wouldn't.
For me, this activity of ours is about having fun. That is the primary goal. Saying that you will not modify an encounter mid-flow to make the game more fun for people contradicts that goal.
Utter hogwash. When you play chess or some other board game, do you play badly on purpose to make the game more fun for the other player? Of course not, because you recognize that having fun in that situation depends on all parties involved trying their best.
I'm saying that you will not modify an encounter mid-flow because I've learned that in the long run that is what makes the game more fun, not because I'm trying to deny fun to the players.
But if you could modify encounters so that everyone has more fun, and you choose not to, then it's you deciding that your principle is more important than your players' fun. And for me, there are very few times that can be the case for any principle I use in GMing.
The problem with that assertion is that relies on two falsehoods. The first falsehood is that the GM is omniscient and so can tell at any point whether modifying the situation does result in more fun. The GM isn't omniscient and can't predict accurately how things are going to go if he changes or doesn't change the situation, and so the real situation you end up putting yourself in is fudging the fudging to get the outcomes you already imagined and wanted. If you decide to at all times freely change the outcomes so that you are omniscient, you are no longer playing to find out what happens. You are just letting the players experience your story. And the second falsehood wrapped up in that is that your story is the one that is most fun for the player. If you are always protecting the outcomes from being anything but you envisioned them to be, well sooner or later the players are going to figure that out and it will harm their fun to know that encounters will shrink or grow depending on how well they are doing, and then also you aren't letting the game or the players surprise you which again harms the fun.
So yes, there are a probably a few times where fudging the game is best for the game for whatever reason - inexperienced players, your mistakes as a GM, terrible luck, etc. - but in my experience if those situations aren't very rare, well the game has problems with its level of fun to begin with.
