I don’t really like running Dungeons & Dragons 5e. There are a lot of reasons for this, but I could learn to live with or subtly work around most of those reasons. There is one core mechanic in the game, however, that drives me up the wall like nothing else, and I basically never see it discussed anywhere—the long rest.
This article is my attempt to make a case for why the long rest isn’t very good. The TL;DR: it encourages a very awkward pacing of the narrative with little flexibility to pace the narrative any differently.
The basics
Before anything in this article can make sense, I need to lay some foundations; four facts that, when combined, gesticulate at a flaw in the mechanic.
First, D&D 5e has combat and attrition as core game mechanics. Your adventurer enters a dungeon with a lot of hit points, spell slots, and potions, and as they delve deeper into the dungeon to fight monsters, your adventurer slowly loses those resources. This resource loss is expected, and is a game unto itself. It’s a balancing act—do I cast fireball to make this fight against a horde of goblins easier, or do I save the spell slot for later and instead risk losing some hit points? Either way you’re losing some resource, and the ‘game’ is to do this in the smartest way possible.
Second, D&D 5e is more fun when you engage with the attrition mechanic. If that fight against the horde of goblins is the only thing happening between now and getting back all your resources, then nothing is stopping you from throwing every resource you can expend at the challenge. This is called ‘going nova’ or the ‘five-minute adventuring day’, and it makes the encounter trivial, which is fun once or twice, but the game is much better when you’re biting your nails to snatch victory from the jaws of defeat. Moreover, the lack of attrition is more beneficial to characters that have rare and powerful resources to expend (wizards) compared to characters that have no such powerful resources (fighters). The intended trade-off is that wizards are weak after their spell slots are gone, and fighters stay at a reliable medium strength come what may. But if the trade-off never happens, then wizards are just flatly stronger, which isn’t very fair and fun. Rather, D&D 5e has the stated assumption that a group of adventurers do six to eight encounters on a day.
Third, D&D 5e isn’t very good at creating balanced difficult encounters. That is to say, you cannot reliably create a single encounter that is difficult-but-survivable for a party that is willing to go nova against it. This is hard to substantiate, but my experience is that the party either exceed expectations and mop the floor with the opposition, or get shown every corner of the damp dungeon cellar in a humiliating defeat, and it’s impossible to estimate in advance which of the two outcomes it’s going to be. Balancing towards the sweet spot—the party prevailing by the skin of their teeth—is like trying to hold onto a wet bar of soap by squeezing it tightly.
Fourth, resources are recovered by in-fiction time passing. Your adventurer can rest an hour to get back some resources in a short rest, or rest for eight hours once per day to get back all resources in a long rest. The party are expected to do frequent short rests.
The dungeon dilemma
If you’ve ever been a game master, you’ve run into the following problem: the party are traversing a dungeon and have lost a good chunk of resources. They do not know what lies ahead or how far it is until they have reached their destination. And they are scared. With so many resources missing, future challenges might just overwhelm them. So, they rightfully reason, this dungeon-delving business would be a lot easier if they had all of their resources at their disposal. They talk it over among themselves and they agree—they should do a long rest. They could barricade a room or simply leave the dungeon to camp outside, and resume their adventure after eight hours of sleep.
And as a game master who cares a little bit about verisimilitude, you are now not really sure how to proceed. What happens in the dungeon while the party are resting for so many hours? It wasn’t so difficult to handwave that the denizens further into the dungeon stayed in their own rooms while the party were steadily advancing room-by-room, but do they now stay in their rooms for eight consecutive hours? If they do not, how do they respond to learning that they have been invaded by a party of adventurers? If they are mildly intelligent, should they not mount a powerful counter-offensive against the party? But if you decide that the denizens act like rational actors that gather arms to stab the party in their sleep, then you are kind of breaking the game. You aren’t supposed to so heavily outmatch the players, and such massive fights aren’t even that fun to run, and there is so much cool stuff deeper in the dungeon that you are excited to show the players.
There are two broad responses to this scenario, and they both suck in their own ways, and have their own ramifications on the game.
You could decide to run the counter-offensive anyway. This is effectively telling the players ‘you lost’. If the players are smart, they try to flee the attack. If the players aren’t very smart, then their characters die or are captured. In any case, after a tedious-to-run encounter, the dungeon delve is over. That’s fine, the players don’t need to win every time, and maybe next time they won’t rest for eight hours when they’re halfway through the baddy’s lair. But then you realise that if you discourage players from doing long rests while dungeon delving, your future dungeons must all be completable without long rests. This limits your creativity a little, because going even slightly over the limit of what you estimate the party might be capable of would mean that the party entered a dungeon that is impossible for them to win, which isn’t typically how the game is played.
The other response is to do nothing, or to ambush the party with a small group of baddies. Because the party are expected to successfully defend against the ambush, and because the party will get back all of their resources after the ambush, these two responses are effectively identical, and the ambush contributes little to the story and the game other than the fun of running combat per se. This may require some additional suspension of disbelief—why would the enemies not strike at the party while they’re weak?—but sometimes, this response is the narratively correct option to take. It makes perfect sense that the undead denizens of a long-undisturbed catacomb stay in their own rooms indefinitely. But if this is your response to the dilemma, you realise, there is nothing stopping the party from going nova against every encounter, advancing through the dungeon at a snail’s pace as they take a long rest after every other encounter.
More veteran game masters have already figured out a more workable solution to the dilemma—add time pressure. If you spare the party a single long rest before something bad happens that they don’t want to happen, then you don’t need to sic the entire dungeon on the party in a counter-offensive (although this may still necessitate a suspension of disbelief), and you disallow the party from going at a safe and boring snail’s pace.
Dungeons & Dragons but without the dungeons (or dragons)
Since the recent boom of popularity of Dungeons & Dragons, people have begun playing the game differently. Where Gygaxian grognards play the game to steal loot from the dragons in the dungeons of an antagonistic game master, newer players are far more interested in crafting collaborative stories in a fantasy world. The players of Critical Role, a widely popular streamed campaign, play like this. The combat becomes a little mini-game that is played between the actual stuff that drives the story, or it becomes the grand resolution of a story arc. And bizarrely, it turns out that dungeons do not make for good locales where stories happen.
Consequently, combat is rarer. Outside of the context of dungeons, it’s simply more difficult to concoct fictional reasons why combat is the correct answer to a wide host of plots. Or, even if combat is the right answer, it’s difficult to justify separating fights into distinct encounters. Fighting a crime boss at the tavern has one, maybe two, or three if you’re stretching it, stages. And unless the fictional schedule is overfull, ‘fighting a crime boss’ seems like a dramatically appropriate amount of conflict before tucking into bed.
What I’m getting at is this—five-minute adventuring days are a lot more sensible in many stories. There is no dearth of instances of one-fight-on-a-day in the books I read and the series I watch. But as asserted earlier, D&D 5e just isn’t very good at supporting five-minute adventuring days. It’s not playing to the system’s strengths.
You could, again, pull out ol’ reliable—time pressure. The crime boss fight cannot be the only thing happening on that day, because her reliable ally the evil witch will do bad things to the party if the party do not also quickly deal with her before she finds out what has happened. And the evil witch has a tower or a lair, obviously, which lends itself well to separated encounters. Between the bar and the lair, you can easily hit the quota of six to eight encounters. Problem solved.
But fabricating reasons to extend the adventuring day becomes both tiring and tired. If you’re on a boat and a kraken attacks, that seems like enough adventuring for one day. This goes for most encounters during travel, in fact. But in D&D 5e, if you want to avoid the five-minute adventuring day, you have to cram in more encounters to balance things out. Do all these additional encounters help drive the plot? Not really, unless you put in more work to connect them to other story events. Does it make sense that they all happen on the same day? Not really, unless you contrive some reason for that. So that day the party are also raided by a bunch of pirates, and they are also ambushed by a group of sirens, and it’s now your job to retroactively justify why these things happen.
Running D&D 5e then becomes like a strange puzzle—how do you bend the story and its beats to match up to the attrition play? The parameters are clear: the party must not be allowed to rest for eight hours, but they must be in a scenario where it makes sense that they can repeatedly sit still for an hour for their short rests. Ideally there are six to eight encounters between long rests, and the encounters should all be of a tailored difficulty. Too easy is boring and doesn’t whittle at the party’s resources, and too difficult ends the game, which is not desirable.
Can I invent stories that fit neatly within this awkward goldilocks zone of fictional urgency? Yes, with a little work. And if you search online for advice on the five-minute adventuring day, that’s the advice you’ll get—put in some extra work, and make sure to add time pressure. But something is lost here. Namely, all the stories that do not fit into the goldilocks zone.
Sometimes—ofttimes—I really just want to exclusively run the kraken attack.
Oberoni’s solution
There is another solution to the goldilocks problem that increases the diversity of stories that can be told. You can use the poorly named Gritty Realism variant rule from the Dungeon Master’s Guide. Under this rule, short rests take eight hours, and long rests take seven consecutive days. You can now do the kraken fight today, and run the pirate raid or something else plot-relevant a few days later, reducing the need to cram everything into a span of 24 hours.
But this displaces the problem. Now suddenly, the types of stories that were viable in the previous goldilocks zone are a lot more difficult to pull off. And if we’re zooming out in time anyway, this new rule supports adventures that take place across days, but not across multiple weeks. Or inversely, we could zoom in instead. Many action films have a lot of plot and action happening in very little time, which is unsupported even by D&D 5e’s default rules.
‘Just pick a time scope and be content with it’, you might say, but I don’t really like that. Different story beats require different levels of intensity across time, and I’d like it if the system were capable of dealing with that.
So if Gritty Realism only displaces the problem, then it’s homebrew time. There is a decent number of possible homebrew solutions to the problem. You can tweak with the times required for rests even more, or even make them variable depending on where the adventurers are. One common homebrew is to exclusively allow long rests in safe locales, typically meaning towns and cities, but this both displaces the problem and doesn’t solve it for adventures that take place in urban environments. You can also reduce the amount of resources restored by short and long rests, but this doesn’t solve the problem of fully rested parties being able to go nova, increases busywork at the table, and also only displaces the problem into a different time scale.
And, ultimately, ‘you can homebrew a solution to a system’s problem’ is a rather tepid observation. I’m not a game designer, and I’m talking about the game as it is, not as it could be. And as it is, D&D 5e just doesn’t do what I want. It forces the stories I can tell into a very specific set of parameters, and if I stray from these parameters, the gameplay suffers thusly that I’d rather be playing any other TTRPG.
Other games
… So I play another TTRPG. At present, I mostly run Savage Worlds. The rhythm of the game is fairly similar to D&D 5e, but the goldilocks problem almost completely evaporates because Savage Worlds simply doesn’t have a lot of resource attrition gameplay. There is some, of course, but not nearly to the same degree as D&D 5e, and resources are a lot easier to recover. Furthermore, it has rules for resolving combat without simulating the entire fight, which can speed certain things along as appropriate.
A game I’d love to be running but that my players refuse to play with me is Fate Core, which solves the problem elegantly both by not having immensely much in the way of resources, but also by tying the recovery of those resources to plot beats. You recover resources when certain things in the story happen, which scales across time in any which way you like. If a lot of things happen in a single day, then you rapidly go through scenes, scenarios, and arcs in that time span, recovering resources whenever these things end. Conversely, an arc could also take months, but contain the same amount of scenes and scenarios.
13th Age has a solution that is similar to Fate Core and that might even make for a good homebrew in D&D 5e. The mechanical equivalent of a short rest happens instantaneously after finishing every second and fourth encounter, and the equivalent of a long rest happens after finishing every sixth encounter. You can space these encounters as far apart or as close together in time as you want, and it won’t make a difference to the mechanical pacing.
Pathfinder 2e, in many ways incredibly similar to D&D 5e, solves it simply by being absurdly balanced. The five-minute adventuring day is a completely viable way to play that game if the encounter is balanced accordingly. Furthermore, hit points are a lot easier to recover in Pathfinder 2e, effectively removing it from attrition play.
It hurts me that D&D 5e is so uniquely bad in this respect, especially because so many people are using it for a style of play that is absolutely not playing to the system’s strengths. It doesn’t need to be like that, and the upcoming One D&D revision appears to be doing all of nothing to address this problem.
Or maybe I’m the only one truly bothered by this.