So how do you guys with my own sentiments (or at least some sympathy for my sentiments) handle these things. What house rules have you developed? Is the game salvageable for someone like us?
I have similar sentiments regarding D&D (I'm more flexible for certain other systems), and I'll say that you can't really totally get rid of it in 5e. However, overall I just think 5e is the best D&D, so I'll tell you what I do (or think could be done) to minimize the problematic elements.
I'll start with your examples and go from there.
1. The player chooses the number of hit dice to apply towards healing during a short rest. There seems to be no analog for the character. There also seems to be a resource being consumed but what is that resource? Potential healing?
The best way is to look at current hit points as your ability to soak up immediate trauma, and HD as representing the rest of your hp. Basically, you can only handle taking about half of your total resiliency's worth of damage in a short period of time, but if you rest up, you can get back into the action.
To minimize metagaming, there are two possibilities. The first is to just say that you have to use HD as the first opportunity. You might make an exception for when you are within 1 or 2 hp of your maximum, but otherwise you have to use them.
The second (and compatible) option is the use the optional Healer's Kit dependency rule from the DMG (must use a healer's kit to spend HD during a short rest). That means spending HD represents the benefit you get from bandaging yourself up and applying ointments and such.
I also use Slower Recovery (puts healing on par with say 3e, or even a bit slower).
So the character chooses to pull out the healer's kit and bandage up, and that gives them a bit more strength to get into the fight. They can only benefit from so much of this before they are worn out and need more than bandages (ie, long rests).
Overall, I haven't found my players using HD unless they plan to get themselves at least close to full, and the kit dependency helps with encouraging that style I think.
2. Action surge. Why is this limited (besides game balance) early on to once between short rests? Can a fighter really only once in the course of a battle choose an exact moment to make an extra effort and then not again? This again seems like the player is choosing something the fighter would know nothing about.
3. Second Wind. A player decides to give his character a surge of energy. The character just gets it apparently unexpectedly. It happens in the fast and furious furer of combat so it's not even something the character could think about much.
I don't think it's
too much of stretch to say they can only pull these off occasionally and that they might be choosing the moment of when to do them. But it is more difficult to explain than the HD, and no optional rules are provided.
I just go with it, and it works fairly well as an in-character choice.
Mechanically, however, Second Wind can be problematic because a fighter can just sit around for a few hours doing it repeatedly and recover all his hp (though not HD). I've house-ruled that a character can benefit from no more than 4 short rests per day, primarily to address Second Wind, but it also works for other things that might be problematic.. So far, we've never run up against a situation where we would exceed those 4, and it does fix the conceptual problem.
You didn't mention Battle Master maneuvers, which are similar to Action Surge and Second Wind. You have a single pool of Superiority Dice you spend on the maneuvers which replenishes on a short rest, so at least it isn't by maneuver recharging. By the book, you don't have to declare using the maneuvers until you hit (or it otherwise comes up). That is simple to fix however. You just say the player has to declare what maneuver they are attempting, and if they fail to hit then the opportunity to complete the maneuver just didn't occur, and hence the full effort wasn't expended (so the die isn't lost). Battle Master's are masters of this stuff, so they know when not to overexert themselves. Of course, you could go hardcore and say the die is expended whether or not the maneuver succeeds, but I think that's unnecessary. The simpler method rarely changes the way the way the game goes (or balance) to any noticeable degree.
In fact, I've decided not worry about it and just let the player declare after success...but my player rarely remembers that and usually declares before taking the action anyway.
Another potentially problematic one is the rogue's Sneak Attack, which you don't have to declare until after the attack hits (though based on other rules, I do believe the intent is that you do declare before you roll damage, thus not allowing you to see if an opponent dies before deciding to use Sneak Attack, so there isn't that particular problem) but you can only use 1/turn. However, when they fail on the first attack and get to try again on the second one, that's where it is weird.
One interpretation is to say that the rogue can spot when the opportunities are there to make a hit be a really good hit. You could say the player has to declare that their character is going for a Sneak Attack before the roll, which shouldn't change how things play out.
Another interpretation of why this is once a turn is that it takes more focus than you can consistently pull off when you are stabbing quickly. I can relate to that general sort of restriction from playing action video games. It really does take more focus than I can maintain to play "at my best" every moment. It doesn't work as well as the first, but works well in combination with the first.
Sneak Attack in 5e works against everything, so it needs to be interpreted a bit differently anyway.
I don't worry about requiring a pre-declaration on this one. I assume the character is always trying to make the best hit they can (ie, they use Sneak Attack at the first opportunity--which my players actually do, so it works), and they just can't maintain the focus on every attack.
4. Inspiration. Since this part of the game is pretty optional (and my guess is anyone close to my thinking ignores it anyway), it's not that big a deal.
Yeah, it's definitely unapologetically metagame. Ignoring it works perfectly fine.
Overall, 5e design seems to have embraced more metagaming than most D&D, but not really done what 4e did. Or at least when they borrowed ideas from 4e in those regards they attempted to tone it down so you could creatively interpret it (like I described above).
If you can work with the sorts of explanations I gave above, you can probably work with pretty much all of the metagame elements in 5e without a serious problem.
Probably one of the biggest things that bugs me is how learning cantrips works. You are limited to only ever knowing a small number of what are supposed to be the simplest spells, even when you can know everything else on your class list automatically (clerics/druids), or theoretically learn them all, including
wish (wizards). Drives me crazy. My house rule is that prepared casters can use a spell preparation slot to prepare other cantrips (which means wizards can add them to their spellbook like any other spell). Prepared cantrips can still be cast at-will. It's not really a problem with casters who know a limited number of spells anyway, though I would allow them to switch a cantrip when they level up, since they can do so with spells of other levels.
This one bothers me so much from a suspension of disbelief perspective that if I'm playing a prepared caster in someone else's campaign I would actually be so bold as to
request my house rule be in effect. I'm not sure I could stomach playing a wizard otherwise.
In practice, no one has ever used my house rule for that. Spell preparations are so precious that my players just haven't wanted to take up a slot with a cantrip, even though it would add an at-will spell to their complement.
I haven't looked at Pathfinder 2 yet. Too crunchy for me, though I'll probably at least scan the SRD to look at the design elements.
