What I took from what
@Ruin Explorer said about initiative being super RNG heavy is in 5e the difference between a low dexterity and a high dexterity is less impactful on initiative order because you're rolling a d20. It's pretty easy for a Rogue with a 20 dex to roll a 1 and go after a character with a dex of 10 because they rolled a 12.
Meanwhile in 2e, even if you're rerolling initiative each round, it's unlikely a character swinging a two-handed sword is going to go before a character using a dagger because the initiative roll is on a d10 and the weapon speed modifier difference is so large. There's still a chance, it's just less impacted by RNG and more by the decisions made by the player to select a particular weapon type.
That's assuming you use weapon speed modifiers...
No, Lanefan, it's not. He was explaining a contrast in rules between editions, to help you understand what Ruin Explorer was talking about in terms of 5E initiative having higher RNG. By contrasting by the book 5E initiative with by the book 2E initiative.
No part of that explanation requires assuming that the reader plays any particular edition. Heck, even if you played 2E WITHOUT all the weapon speed modifiers it would STILL be less random than 5E because it uses a d10 instead of a d20, so the standard initiative modifiers (and any optional ones, like weapon speed and spell casting times) have a proportionally greater impact.
One thing I don't think is actually a problem is the "expected" 6-8 encounters because it isn't actually expected. All I've ever been able to find about this amount of encounters is in the DMG and all it is saying is that a typical party can handle around 6-8 medium or hard encounters, it never says that it's an expected or required number. People seem to have taken that as some sort of gospel to the point where some will say people are playing wrong if you don't have that number of encounters, something like "X DM doesn't run 6-8 encounters a day and complains that they can't challenge their PCs

".
Well, sure, it's the maximum not a norm. But there has been a strong correlation for a decade between people complaining about balance, who when they provide details admit they are veering way off the book expectations. If the players aren't pushed, they won't be challenged in the resource management game...which isn't necessarily a problem. But not challenging a party, and then saying the game doesn't provide a challenge...
Right. That number of encounters is expected to challenge an average party, where "challenge" means to drain their resources sufficiently that getting through those encounters doesn't seem easy or a pushover.
And I think that this is a useful figure if you're playing a dungeon crawl, but it's much less useful if you're doing other things, where a slower pace of encounters makes more sense.
But what if the dm doesn’t want to run a game that’s all about long days of battle? What if they want intrigue and exploration with an occasional fight?
5e doesn’t support that, nor are there easy houserule options to achieve it.
I mean, the Gritty Realism optional rule works pretty well. Overnight short rests and long rests which take a week. That allows for a slower encounter/resource recovery pace.
ok, i know you're joking, but unironically pf2e does this and it (along with the game's math, obviously) genuinely helps clarify how powerful an average PC is compared to a particular monster.
PF2 is modeled on 4E D&D in that. One of the sacred cows 4E slew for clarity was the spell level vs caster level disparity/ratio. In 4E a 5th level character can cast 5th level spells and use 5th level other powers. So Fireball was a 5th level Daily power, for example.
I’d rather modules like that be DM-facing, but if it’s gonna be player side (and nothing inside the class should be DM-decides. Ever.) I’d want it to be very much fully player side.
No “optional variant ask your DM”. No.
It’s gotta be “When you gain second level as a Paladin, you can choose between [long rest casting model], [short rest casting model], and [short and long rest mixed casting model].
But really for that part I think a better idea is to give every class some things that recharge on a long rest and some on a short rest.
I think the better approach would be to set one version as the default in the PH, for simplicity and consistency on what the baseline expectation is, then put the variants at the end of the class description or in the DMG.