It boils down to closer to a short rest every other encounter on a 6-8 day, or after each encounter on a 3-4, but, yes, with symmetrically longer & more challenging encounters that would seem to make sense. One factor that could mess with the idea, though, is whole-encounter effects.
Yes, you're right. When I said a short rest whenever I was thinking whenever the players like, which will be after almost every encounter when your adventure model is 3-4 deadly+. The important variable for balancing short and long rest classes is the ratio between short and long rests (assuming equivalent resource depletion).
That makes very little sense, even as rationalizations go.
[snip]
It's a natural-language interpretation, and 5e favors natural language over obscure jargon - and, I should hope, over intentionally deceiving both players & DMs.
I don't find it hard to believe that the designers would want to let the DM know how many encounters the typical party can get through while knowing that this won't happen in practice most of the time. To get the number of encounters the PCs do get through to consistently align with the number of encounters they can get through requires rest restriction (of the sort @
Flamestrike advocates), which is not a trivial modification of the game and the DMG doesn't give an advice on how to do that. So I find it hard to believe that the designers expect that to occur.
In the typical adventure the players get a "free" rest at the end--when you clear the dungeon and go back to town you rest whether you need to or not. So if the DM prepares a 6-8 encounter dungeon, and the party takes one long rest during (say when they feel that the "boss fight" is coming up) and one after completing it, that's like 5 or 6 encounters before the first long rest, and then 1 or 2 tougher ones at the end. I'm sure they realize that's how the typical adventure plays out.
But, it's none of it set in stone. If you have a party where one PC seems to be underperforming, and his resource schedule is the odd man out, that could be a place to look for solutions, for instance. OTOH, if you have one mostly-daily character dominating and another languishing, changing encounters/day around is unlikely to help. There's a lot of ways the DM can tweak his game to keep it fun for everyone. Encounters/day is one of them, and it's one that tweaks several things at once - including intraparty class balance based on resources, and encounter difficulty - and not always conveniently in all the directions you need.
Yes, trying to control encounters per day is a clumsy way to troubleshoot the game. I think in most cases it's better to encourage the party to take more short rests than to try to prevent them from taking long rests.
Over 3-4 encounters with 1 short rest:
That changes the ratio of short to long rests. If you run the numbers comparing 6 encounters with a short rest after every second vs 3 encounters with a short rest after each (same 2:1 ratio of short to long rests), you will find the classes are balanced similarly.
There are a few additional considerations: long rest classes have more AoE abilities and (slightly) weaker basic attacks, which is an advantage with fewer encounters, and with longer adventuring days they're also more likely to be too conservative using their abilities, "wasting" them more often. So I would expect the short rest classes to come out a bit ahead when you look at damage inflicted and healed with 2 short rests per long rest to account for that. If not, I would admit that implies the game is balanced to some extent in favor of longer ADs.
No-one is saying it's a 'hard' limit mate. Its a guideline that you build encounters around (and you can frequently ignore it or intentionally tinker with it if you want). You dont cram 6-8 encounter adventuring days down your parties throats day in day out. You mix it up, limiting some adventures, leaving it in the parties hands for others, pushing shorter ADs on the party from time to time, and pushing longer ones on them from time to time.
Planning encounter by encounter is too on-the-nose for me. I prefer the sandbox approach where you leave the choice of when to rest in the hands of the players but add more variables and greater variance into the encounters and the environment to make it a more interesting choice. My focus is on giving the players interesting decisions to make, not depleting the PCs' resources. A game that's hard on the PCs is not necessarily hard on the players.
I don't like to plan for variety in general. That's inefficient. If more variety would make the game better you should use more randomness in your adventure design procedure.
And as for spells that allow the party safe rest (rope trick etc), they are all well and good. [snip]
Yes, I don't disagree with the rest of your post. If you ramp up encounter difficulty to the point where you get 5MWD issues then you do have to start restricting rests, either with hard limits or stiff disincentives. I'm pointing out that the DMG doesn't advocate this, and by default rests are only very mildly disincentivized, not to argue that restricting rests is a bad idea but to cast doubt on the notion that the statement that the party has enough resources to last 6-8 encounters constitutes advice that the game plays best with 6-8 encounter days.
There's nothing magical about 6-8 encounters/day. The magic number, to the extent that there is one, is 2 short rests per long rest.
The game seems designed to play well with a variety of encounters per day.