So if it's from many patches ago why do you think it's the be all end all authority on the most effective way to play the game when it's over a year old?
The game has changed in multiple ways since then and I suspect the streamer was using unintended exploits of some kind too.
Either way it's a moot point because if the best way to play the game is to ignore half the game mechanics, that should be changed too.
I think it is optimal because it was the fastest I'd ever seen and not exploiting any bugs that have since been patched out.
The early game towns still have limited fortification levels and defensive buildings at the game start. The starter garrisons are still small. The campaign AI has been made much dumber, if anything. Recruiting works the same way, although it is easier/more common to get high-tier troops. The AI parties still drop off troops in every settlement you enter. Fresh parties can still be called to armies. Army cohesion can still be topped up using influence. Players can still take troops from their clan parties. The limiting player resources (food, influence, gold) have a similar economy or the same. Sieges have gotten easier since they fixed ladders and towers.
Plowing through settlements without giving a **** about your troops isn't exploiting anything, just mastering the various systems already in place -- open recruitment, early game fortification/garrison weakness, player mana generation, AI decision-making and live sieges favoring the attackers.
If I wanted to link an actual
speedrun, filled with the usual speedrun-style exploits, I could've just linked Hcaylion or Lesser Scholar's six and seven hour WC speedruns. But both of those very obviously and clearly abuse outright bugs (getting Trade up to 300 in a few minutes), whereas just being the sort of person who literally doesn't give a damn about his troops in M&B is only counterintuitive, not an exploit. I sure as hell didn't figure it out on my own, despite getting hints from Bannerman Man.
Bannerlord doesn't need to be speedrunner-proof, but its mechanics and different troop types should at the very least not be waving the fact in your face that there's no point in doing them and you're operating at half efficiency. Or that your elite troops kill everything effortlessly before your regular troops can even get in range to fight.
That would involve a complete overhaul of Bannerlord's combat.
edit: As for the bit about elite troops killing everything is working at crosspurpose with the desire to make troop differences meaningful. In battles ranging in size between 1-1000 troops that are divided by a grab bag of dozens of types, only huge differences in effectiveness are noticable. You can tell the difference between shielded and unshielded, well-armored and basically naked, on horse or on foot, ranged or melee, sure. That's plenty obvious. But you need sterile room, 500v500 tests to see if one type of similar unit is better than another, which is something that never happens in the campaign, and can be "broken" by the player simply issuing one or two intelligent commands.
What a terrible idea. Hopefully you were being sarcastic. Making every T4-T6 troop use the same weapons would be very boring.
I was being completely serious when I said that if noble troops are going to be rare and desirable for players, they need to be powerful, and not just, "Oh, here's a slightly better horse archer or lancer." Because outside of aesthetics or RP, there is no reason to care by time you're in a position to afford them if that's the case.