D&D 5E Resting and the frikkin' Elephant in the Room

[Tony Vargas]

I look at it like this. In order to satisfy the OP's needs, we'd have to rewrite the entire game.

I don't think so. I can imagine a couple of ways the DM could rule to fix things up without actually authoring a lot of variants, and some mere wording changes that could make the game more open to such interpretations....

The entire Monster Manual would need to be rewritten in order to make individual monsters more of a threat so as to satisfy his need for single encounter days.

Not one of the approaches I was considering, but not as hard as it sounds - simply overhauling CR numbers could do probably be sufficient on that end.

Additionally, classes would need to be rewritten so that every class balances over the course of individual encounters, instead of over the course of a day.

That is a major stumbling block to a purely mechanical fix, yes.

IOW, we'd need to rewrite 5e to use 3e style monsters and 4e style character balance.

Considering what 3e-style monsters were like, that doesn't sound like a fantastic idea. DMing is enough work without monsters as detailed and system-mastery-requiring and optimized PCs.


[sblock="...hopefully the last..."]

Umm, Imaro, Essentials classes focused a LOT more on At Will abilities than on dailies.

Of course. To be fair, Essentials did re-introduce a pacing-sensitive imbalance between the new AEU martial classes and the traditional AEDU classes, including the powered-up Wizard sub-classes, but it only edged in that direction a little, the Fighter & Thief lost their Dailies, but the Wizard only got more powerful encounters & greater versatility in preparing spells (all but at-wills became preped, and they could retrain new spells without losing the old ones - unmatched Tier-1-prepped-caster-style 'strategic flexibility,' by 4e standards, though still barely a hint of what other editions provide).

Psionic power points recovered on short rests, not long ones. Psionic characters were not balanced over the day.

Well, pre-Essentials, classes were balanced in spite of day length, which, I know, is what you said. I just felt like rephrasing it.

4e, throughout it's run, focused balance on the individual encounter. The encounter, not the adventuring day, was the base unit of all classes.

In terms of class balance, mostly sorta true. 4e classes had rough resource parity, so they were balanced (with eachother, to the extent the designers got the powers right), regardless of day length - all classes were a bit more powerful when they popped dailies in an encounter, for the obvious Elephant-relevant example. But the design still had daily resources, particularly surges, and wasn't designed solely around the encounter (an example of encounter-based design would be the 'D&D Gamma World' system, which was similar in many particulars, but, was, well, all-encounter - powers & even hps refreshed after each encounter). And, you could still design an adventure around attrition over the day, it'd be primarily hp/surge attrition, as the exhaustion of dailies wasn't too devastating, and milestones granting action points and un-locking additional magic-item dailies mitigated it a little.

Monsters were designed around the encounter - since a given monster usually only existed for one encounter, afterall. Since PCs weren't designed solely around a single encounter in that way, encounter balance (difficulty) was impacted by encounters/day. So players could engineer an advantage in one encounter by successfully avoiding earlier ones, for instance.

[/sblock]

While I understand the impetus, I'm also very realistic in that this is just not going to happen. There is no way that WotC is going to do this. It's a pipe dream.

More importantly, few DMs are going to have both the exceptional design talent and the sheer time to pull it off.

So, being the pragmatic person that I am, I offer solutions that don't require rewriting the entire game. Plan your adventures such that you are no longer relying on single encounters to provide an adequate challenge. We did this in 1e all the time since 1e monsters individually compared to the PC's were extremely weak. So, you designed adventures where you had waves of baddies, strung together encounters, and a much more restricted recovery rate.

Acknowledges the elephant, which is good, but, still as has been pointed out, not a solution.

Even so, it's a perfectly reasonable way to run if you like that kind of pacing.

IOW, the solution to the "Elephant in the Room" is to either accept the system as it is and work within that system or find a different system. And, AFAIC, that's the long and the short of it.

That's really not in keeping with the spirit of 5e. 5e is not a take-it-or-leave-it system, it's a make-it-your-own system.
Now, it is impractical for any but the most remarkably capable and rich-in-leisure-time DM to re-build 5e rules to impose mechanical balance that replaces the 6-8 encounter/day guideline, but there are surely other ways to acknowledge and tame that elephant....

On the rulings-not-rules side, the DM can assert the same privilege to make judgments that he has in most other areas of the rules over the length of short & long rests. That's not much, and it's odd, IMHO, that the PHB didn't just do so from the beginning. A few weasel-words ('generally take...' 'your DM may allow...' '..recover of up to,' etc) would have left the requirements and benefits of rests in the DMs court. The default 1 & 8 hr rests could have been spun as 'typical' or 'under ideal conditions' or something. No actual rules would need to have been changed, just presented more like, well, so many other rules in 5e.
Using that latitude the DM wouldn't need to fit the pacing of his campaign to the 6-8 encounter guideline, he could, instead, fit the requirements/benefits of resting to the campaigns' pacing & situation at the time. In a long, difficult journey, rests are impossible for days & weeks on end. In an intense high-adrenaline raid, even a few minutes to catch your breath counts as a short rest. Whether the DM presents that arbitrarily, in a linear structure or like an old-school-wargaming scenario-specific variant, or comes up with such rulings in response to player choices & situations, it should be perfectly workable.

One the 'make the game your own' side, of course, it's harder to do at a stroke, entirely with mechanics, but if you do want both class balance and a consistent pacing that is different from the 6-8 encounter day, the DMG does have a modules that change the rules around rests, and just changing those those rules if no module is quite right for the desired pacing, is a much more practical undertaking than re-writing all the classes. The key is consistent pacing. You can make rests take more less time to fit a campaign that's generally going to be chugging along faster or slower than 6-8 encounters/day, but you're really just re-defining 'day,' and conforming to the guidelline.

 

[clearstream]

the DM can assert the same privilege to make judgments that he has in most other areas of the rules over the length of short & long rests. That's not much, and it's odd, IMHO, that the PHB didn't just do so from the beginning. A few weasel-words ('generally take...' 'your DM may allow...' '..recover of up to,' etc) would have left the requirements and benefits of rests in the DMs court. The default 1 & 8 hr rests could have been spun as 'typical' or 'under ideal conditions' or something. No actual rules would need to have been changed, just presented more like, well, so many other rules in 5e.

I agree with you here albeit I would contextualise this approach as having being available to all DMs over all versions of the game That said, I feel like one of the advances of 5e is the use of more natural language with less attempt at precision. As you say, it's odd that wasn't done with the wording of rests.

Using that latitude the DM wouldn't need to fit the pacing of his campaign to the 6-8 encounter guideline, he could, instead, fit the requirements/benefits of resting to the campaigns' pacing & situation at the time. In a long, difficult journey, rests are impossible for days & weeks on end. In an intense high-adrenaline raid, even a few minutes to catch your breath counts as a short rest. Whether the DM presents that arbitrarily, in a linear structure or like an old-school-wargaming scenario-specific variant, or comes up with such rulings in response to player choices & situations, it should be perfectly workable.

I think so too. I wouldn't be so arbitrary, but management of the pacing does the job quite well. In tonight's session my PCs carried out an extended series of tasks and travelled 5 further days before resting, which I found satisfactory.

One the 'make the game your own' side, of course, it's harder to do at a stroke, entirely with mechanics, but if you do want both class balance and a consistent pacing that is different from the 6-8 encounter day, the DMG does have a modules that change the rules around rests, and just changing those those rules if no module is quite right for the desired pacing, is a much more practical undertaking than re-writing all the classes. The key is consistent pacing. You can make rests take more less time to fit a campaign that's generally going to be chugging along faster or slower than 6-8 encounters/day, but you're really just re-defining 'day,' and conforming to the guidelline.

Indeed. I firmly believe that interest is maximised through appeal to both chains of encounters and single deadly encounters. And that 5e classes are significantly balanced around a meaningful mechanical difference between short-rest and long-rest availability. In the simplest terms, around 6-8 encounters between long rests, although that should be understood as highly flexible. As noted, the Warlock exemplifies that balance with limited spell slots that all recover on a short rest.

 

[Lanefan]

Ah, I get it. In the October Sage Advice compendium I see this question and answer

"Can you also cast a reaction spell on your turn? You sure can! Here’s a common way for it to happen: Cornelius the wizard is casting fireball on his turn, and his foe casts counterspell on him. Cornelius has counterspell prepared, so he uses his reaction to cast it and break his foe’s counterspell before it can stop fireball."

What is happening here is that Cornelius is using his Action to cast Fireball, and his Reaction to cast Counterspell. Casting a spell doesn't require uninterrupted effort. No timing finesses of the MTG sort are needed.

Well, actually, timing finesses like MtG are needed in order to allow the in-game fiction to make any sense whatsoever.

First basic assumption: every casting of a given spell pretty much works the same. When you cast Fireball it's just the same as when I cast Fireball...it takes the same amount of time, we go through the same motions, and we produce a ball of fire somewhere nearby. All agreed?

Second assumption: some spells are quicker to cast than other spells. Counterspell, for example, is quicker to cast than Fireball (reaction vs. action). All agreed?

Third assumption: successful interruption of a spell normally causes that spell to fail or fizzle or whatever term you like. All agreed?

And so, given those assumptions, let me introduce you to the rabbit holes. There are several.

Rabbit hole #1 - interruption. To cast Counterspell I by definition have to interrupt the spell I'm already casting...I can't normally cast two spells at once, and a counterspell effect isn't a built-in feature of a normal Fireball...and so my Fireball should be lost either by my own interruption of it or by it being countered by my foe. But the rules say it resolves...wtf?

Rabbit hole #2 - timing. If any given spell is always the same as in assumption 1 above, if my enemy's Counterspell starts first then - even if we allow me to react to it and interrupt myself to cast my own Counterspell - then his should and must resolve before mine does, as mine takes the same amount of in-game time as his and he started casting first. But instead we have last-in-first-out, based on the MtG "stack" rules, which is kinda ridiculous in the fiction - but there's your timing finesse. In short, it should be impossible to Counterspell a Counterspell.

Rabbit hole #3 - concentration. If I can interrupt my concentration on casting Fireball enough to cast another spell and then resume my Fireball, I can by extension then interrupt my concentration at all sorts of other times to do all sorts of other things using the same logic. End result: concentration becomes a much weaker mechanical restraint on casters.

Dumb rule.

Lan-"the sad part is that this would be so easy to fix or errata, but it hasn't been done"-efan

 

[Tony Vargas]

I firmly believe that interest is maximised through appeal to both chains of encounters and single deadly encounters.

I suppose that varies like other preferences.

And that 5e classes are significantly balanced around a meaningful mechanical difference between short-rest and long-rest availability. In the simplest terms, around 6-8 encounters between long rests,

I've been omitting 2-3 short rests because it just gets unwieldy, but the balance point does seem to be, or at least to be intended to be, based on the guidance provided, 6-8 encounters between long rests, with a short rest about every-other encounter. With most encounters not taking all that many 6-second rounds, that means the prescribed adventuring 'day' would consist of 3-4 minutes of encounters, 10-11 hours of resting, 12+ hours for the other two pillars and whatever else you do to kill time until you can take your next long rest...

We may need a new metaphor for the 5MWD.... ;P

Well, actually, timing finesses like MtG are needed in order to allow the in-game fiction to make any sense whatsoever.

Certainly, off-turn reactions and the like are, otherwise turn-based cyclical initiative is even more 'dissociated.' But, just because one way of visualizing the fiction doesn't make sense doesn't mean another can't, even given all the same mechanics...

First basic assumption: every casting of a given spell pretty much works the same. When you cast Fireball it's just the same as when I cast Fireball...it takes the same amount of time, we go through the same motions, and we produce a ball of fire somewhere nearby.

Seems unlikely. For one thing, different casters have features that can change who their fireball works, that probably means they're casting it a little differently. Then there's slots, a 3rd level fireball and a 5th level fireball aren't the same thing, probably they're not identical to cast.

Second assumption: some spells are quicker to cast than other spells. Counterspell, for example, is quicker to cast than Fireball (reaction vs. action).

True in broad cases. Rituals take longer than spells that are cast in a single action, which could be visualized as taking longer than those cast with a bonus action or reaction. (Though, an attack roll can be done, by the same character, with the same kind of weapon, as an action, bonus action, or reaction, depending on what options that character has available, so the action types might not imply greatly different literal times to accomplish).

Third assumption: successful interruption of a spell normally causes that spell to fail or fizzle or whatever term you like.

It's not that easy to successfully interrupt a spell, but, sure, in theory...

Rabbit hole #1 - interruption. To cast Counterspell I by definition have to interrupt the spell I'm already casting...I can't normally cast two spells at once, and a counterspell effect isn't a built-in feature of a normal Fireball...and so my Fireball should be lost either by my own interruption of it or by it being countered by my foe. But the rules say it resolves...wtf?

Alternate visualization #1: You start casting fireball, calling up whatever the heck mystical reserves 'slots' are meant to model, another caster casts counter-spell - the two of you engage in a moment of magickal arm-wrestling and either the fireball goes off or it doesn't. That may or may not include a counter-counter-spell from you - the fiction is, two wizards struggled over that fireball and one prevailed.

Rabbit hole #2 - timing. If any given spell is always the same as in assumption 1 above, if my enemy's Counterspell starts first then - even if we allow me to react to it and interrupt myself to cast my own Counterspell - then his should and must resolve before mine does, as mine takes the same amount of in-game time as his and he started casting first.

Rationalization #2: Counterspells take an amount of time to cast that is always less than the spell they're countering. Thus a counterspell is faster than a fireball, but a counter-counterspell is faster than counterspell, and a counter-counter-counterspell is faster still...

Rabbit hole #3 - concentration. If I can interrupt my concentration on casting Fireball enough to cast another spell and then resume my Fireball, I can by extension then interrupt my concentration at all sorts of other times to do all sorts of other things using the same logic.

Interpretation #3: Concentration applies to spells that take longer than a single action to cast. Spells cast with action or reactions don't invoke concentration, so your rabbit-hole is a mole-hill.

When you cast a spell with a casting time longer than a single action or reaction, you must spend your action each turn casting the spell, and you must maintain your concentration while you do so (see “Concentration” below).

Lan-"the sad part is that this would be so easy to fix or errata, but it hasn't been done"-efan

Tony "Polishing his 5e Apologist Credentials" Vargas

 

[clearstream]

Well, actually, timing finesses like MtG are needed in order to allow the in-game fiction to make any sense whatsoever.

From one perspective that's exactly right, and from another... well, I believe ignoring such complexity is the compelling reason for the ruling being as it is. Setting aside narrative sense for a moment and focusing on the mechanics, I can cast a spell with my action and a spell with my reaction and I simply ignore how those overlap. That avoids a bunch timing finesses, corner-cases and exceptions. It doesn't so much fix those issues, as excise them from the game. Thus, no timing finesses of the MTG sort are needed

Second assumption: some spells are quicker to cast than other spells. Counterspell, for example, is quicker to cast than Fireball (reaction vs. action). All agreed?

COWTRA we know no such thing. What we know about actions is that they happen between the start of my turn and the end of my turn. What we know about reactions is that they happen when triggered. We know that I can only ever have one reaction between the start of my turn and the start of my next turn (aka a round) and we know I can have up to two actions in that same frame of events. What I'm getting at is that the mechanics have been simplified in a way that is cleverly quite "gamey" without overmuch seeming that way. If we try to add back in an idea of a spell not happening in an action (or series of actions) but having temporal extension, then exactly as you observe we end up needing to increase the mechanical complexity.

Rabbit hole #1 - interruption. To cast Counterspell I by definition have to interrupt the spell I'm already casting...I can't normally cast two spells at once, and a counterspell effect isn't a built-in feature of a normal Fireball...and so my Fireball should be lost either by my own interruption of it or by it being countered by my foe. But the rules say it resolves...wtf?

Time doesn't enter into it. Resources do. Think of the caster as a multidimensional entity. They have one zone called "action" that they stick their Fireball in. And another zone called "reaction" that they stick their Counterspell in. The two zones aren't temporally connected although, when played out on a common plane they can be sequenced. Alternatively think of Action as blue candy and Reaction as pink candy.

Rabbit hole #2 - timing. If any given spell is always the same as in assumption 1 above, if my enemy's Counterspell starts first then - even if we allow me to react to it and interrupt myself to cast my own Counterspell - then his should and must resolve before mine does, as mine takes the same amount of in-game time as his and he started casting first. But instead we have last-in-first-out, based on the MtG "stack" rules, which is kinda ridiculous in the fiction - but there's your timing finesse. In short, it should be impossible to Counterspell a Counterspell.

Counterspell does <thing>. Where <thing> is that it targets a spell as it is being cast and stops it resolving. Timing doesn't come into it. Counterspell consumes pink candy: no pink candy means no Counterspell. Counterspell doesn't care about blue candy.

Rabbit hole #3 - concentration. If I can interrupt my concentration on casting Fireball enough to cast another spell and then resume my Fireball, I can by extension then interrupt my concentration at all sorts of other times to do all sorts of other things using the same logic. End result: concentration becomes a much weaker mechanical restraint on casters.

Concentration isn't involved in casting a Fireball. Or a Counterspell for that matter. Neither have the concentration tag.

Dumb rule.

In its defence, it sneakily avoids what we conscientious types know are all kinds of complicated mechanical problems such as we see in the 200 or more pages of precise MTG rules. I believe WotC made a conscious choice to make sure such precise rules didn't creep into 5e (they do in places, but overall they're avoided).

Lan-"the sad part is that this would be so easy to fix or errata, but it hasn't been done"-efan

Von-"you could do so at the cost of a lot more complexity, do you really want that?!"-klaude

 

[clearstream]

I suppose that varies like other preferences.

Bah, game analysis is certainly at the point that we can draw a distinction between better and worse design. But if all you mean is that not all DMs will agree, then that's a given. If you offered some DMs a magical life-giving unicorn they'd argue with you about the teeth.

I've been committing 2-3 short rests because it just gets unwieldy, but the balance point does seem to be, or at least to be intended to be, based on the guidance provided, 6-8 encounters between long rests, with a short rest about every-other encounter. With most encounters not taking all that many 6-second rounds, that means the prescribed adventuring 'day' would consist of 3-4 minutes of encounters, 10-11 hours of resting, 12+ hours for the other two pillars and whatever else you do to kill time until you can take your next long rest...

We may need a new metaphor for the 5MWD.... ;P

For me the 5MWD was a synonym for the 1EWD (1 encounter working day). It's not about the calendar time spent by characters in the furnace of combat, which rightly should be brief. Rather it's about whether their count of encounters between rests is 1 or 2, or 1 or 2 and other values greater than 1 or 2. That's one thing I appreciate from WotC's exploration with 4e. We learned a lot about that problem, even while not solving it. (I know 4e was an exploration into other things, too, seen earlier in ToB. They were also important.)

 

[Tony Vargas]

Bah, game analysis is certainly at the point that we can draw a distinction between better and worse design.

Good luck with that.

But if all you mean is that not all DMs will agree, then that's a given. If you offered some DMs a magical life-giving unicorn they'd argue with you about the teeth.

Some DMs will find their group more engaged with a longish, multi-encounter 'day,' some with a series of all-out single encounter ones, some with a variety. That the game being able to handle all of those well, without completely trashing class balance and expected encounter difficulty, would be a mark of better design than a game that handled only one of 'em, I have to agree.

For me the 5MWD was a synonym for the 1EWD (1 encounter working day). It's not about the calendar time spent by characters in the furnace of combat,

Yeah, I as just having fun with it.

Rather it's about whether their count of encounters between rests is 1 or 2, or 1 or 2 and other values greater than 1 or 2. That's one thing I appreciate from WotC's exploration with 4e. We learned a lot about that problem, even while not solving it. (I know 4e was an exploration into other things, too, seen earlier in ToB. They were also important.)

The way it seemed to me, there were always two distinct sides to the 5MWD problem. Class imbalance and encounter difficulty. By putting classes on comparable resource schedules, you solve the former. By concentrating less of PC power in daily resources, you reduce the swing in the latter (though, depending on style, some swing may be desirable).

5e openly balanced classes and estimated encounter difficulty based on a 6-8 encounter day, so it's not surprising that it has the same familiar 5MWD issues as the classic game. In fact, that guidance as to the number of encounters/day at which the classes would balance was one of the Next Playtest promises.
One that was actually kept.

 

[OB1]

Keep in mind that the 6-8 encounter guideline is only an example of how the specific rule for using encounter and daily xp works. What is important is 1 Deadly = 2 Hard = 3 Medium = 6 Easy such that you can get the following daily work flow giving you anywhere between 3 and 18 encounters and still be balanced for both SR and LR classes.

Deadly-SR-Deadly-SR-Deadly-LR
Hard-Hard-SR-Hard-Hard-SR-Hard-Hard-LR
Med-Med-Med-SR-Med-Med-Med-SR-Med-Med-Med-LR
Easy-Easy-Med-Med-SR-Hard-Hard-SR-Deadly-LR
Med-Med-Med-SR-Easy-Easy-Hard-Easy-SR-Deadly-LR
Deadly-SR-6xEasy-SR-Deadly

 

[Tony Vargas]

Keep in mind that the 6-8 encounter guideline is only an example of how the specific rule for using encounter and daily xp works. What is important is 1 Deadly = 2 Hard = 3 Medium = 6 Easy such that you can get the following daily work flow giving you anywhere between 3 and 18 encounters and still be balanced for both SR and LR classes.

As long as you maintain the 2-3 short rests to the long rest ratio, sure, more or less. There's also the number of rounds to consider, because it's not just SR balancing with LR recharges, it's also both of them balancing with at-will.

 

[Lanefan]

From one perspective that's exactly right, and from another... well, I believe ignoring such complexity is the compelling reason for the ruling being as it is. Setting aside narrative sense for a moment and focusing on the mechanics, I can cast a spell with my action and a spell with my reaction and I simply ignore how those overlap.

Which isn't, in any RPG, something I want to have to do. Instead, I want the narrative sense and the mechanics to agree.

That avoids a bunch timing finesses, corner-cases and exceptions. It doesn't so much fix those issues, as excise them from the game.

It doesn't fix them, nor does it excise them. It just sweeps them under the rug and pretends they're not there; a thoroughly unsatisfactory solution.

COWTRA we know no such thing. What we know about actions is that they happen between the start of my turn and the end of my turn. What we know about reactions is that they happen when triggered. We know that I can only ever have one reaction between the start of my turn and the start of my next turn (aka a round) and we know I can have up to two actions in that same frame of events. What I'm getting at is that the mechanics have been simplified in a way that is cleverly quite "gamey" without overmuch seeming that way. If we try to add back in an idea of a spell not happening in an action (or series of actions) but having temporal extension, then exactly as you observe we end up needing to increase the mechanical complexity.


Time doesn't enter into it. Resources do. Think of the caster as a multidimensional entity. They have one zone called "action" that they stick their Fireball in. And another zone called "reaction" that they stick their Counterspell in. The two zones aren't temporally connected although, when played out on a common plane they can be sequenced. Alternatively think of Action as blue candy and Reaction as pink candy.


Counterspell does <thing>. Where <thing> is that it targets a spell as it is being cast and stops it resolving. Timing doesn't come into it. Counterspell consumes pink candy: no pink candy means no Counterspell. Counterspell doesn't care about blue candy.

But really, how hard would it be to simply say that if either your action or your reaction involves casting a spell you can't do them at the same time? That right there gets rid of just about all the dissociation.

Concentration isn't involved in casting a Fireball. Or a Counterspell for that matter. Neither have the concentration tag.

They don't have the concentration tag to maintain, as both are instantaneous; but I thought all spells required some sort of concentration (or focus, or paying attention, or whatever term) while casting.

In its defence, it sneakily avoids what we conscientious types know are all kinds of complicated mechanical problems such as we see in the 200 or more pages of precise MTG rules. I believe WotC made a conscious choice to make sure such precise rules didn't creep into 5e (they do in places, but overall they're avoided).

Von-"you could do so at the cost of a lot more complexity, do you really want that?!"-klaude

Two very simple and non-complexity-adding rules are all that's needed:

- - First, (repeated from above) if either your action or your reaction involves casting a spell you can't do one at the same time as the other. (so, no reaction spells in the middle of any other action, and no reactions at all in the middle of an action if that action is a spell; but melee reactions during a melee action remain OK)
- - Second, reactions resolve in the order they are declared (so first in, first out).

Rationalization #2: Counterspells take an amount of time to cast that is always less than the spell they're countering. Thus a counterspell is faster than a fireball, but a counter-counterspell is faster than counterspell, and a counter-counter-counterspell is faster still...

Yeah, then we're into horrible MtG-style counterspell battles. Why not just have it that reaction-speed spells (along with any other reactions) resolve in the order declared and avoid all this crap?

Lan-"not a fan of counter decks in MtG either"-efan