Thoughts on Recharge?

[Pooky the Imp]

I don't know if this is an odd question but I'm curious to hear what people think of Recharge as a mechanic for balancing monster abilities?

Do you prefer it to alternatives (e.g. limited uses per long/short rest, the 'wait 1d4 rounds before using it again' version in 3.5, some other mechanic you can name etc.)?

Also, do you like that this type of ability only exists on monsters - never for PCs?

[PhoenixPhyre]

Limited uses per rest on a monster generally means "spam this every turn" unless the limit is 1 or it has specific trigger criteria.

Fixed recharge timers are limited in use because most monsters are around for 3-5 rounds at most.

Dice based recharge means it won't usually happen every round, so you have to mix it up, but there's a substantial change of getting to use it twice in a combat.

So each one has its advantages and disadvantages. Each one has its place.

And I'm 100% ok with it being monster only. And would prefer fewer ways for players to get as access to monster blocks.

One other method might be "recharges on (event)". 4e used a lot of "recharges on being bloodied for the first time in a fight" (ie dropped to half or less HP)

[icedraikon]

I hate recharge mechanics. On the 5-6, generally they're used early in combat like round 1-2, then expected to come up near the end of combat to help change the pace for when the PCs are probably starting to clean up. However, if you get a few recharges in a row it can absolutely **** encounter balance. I prefer the idea of the 1d4 wait # of rounds or something like that. Use the recharge ability and instantly roll the 1d4 so the players can kinda get a feel for how long they have before the ability comes back. Similar to the ancient sun dragon scene in Dragon Prince where you can see the energy coming back. That way, you can play around it and gets players thinking more instead of "oh he rolled is 20ft radius nuke + stun recharge again, nice guess I'll stay stunned for this combat"

[RogueJK]

I hate recharge mechanics. On the 5-6, generally they're used early in combat like round 1-2, then expected to come up near the end of combat to help change the pace for when the PCs are probably starting to clean up. However, if you get a few recharges in a row it can absolutely **** encounter balance.

100% this. It makes the difficulty of certain monsters super swingy (especially in groups), and hard to account for. Too easy for them to end up being either total cakewalks if their recharge dice roll poorly, or wipe the party when they recharge every round.

Mind flayers are a prime example. A small group of mind flayers can absolutely wipe the floor with a party of significantly higher level PCs if they get lucky enough on their recharge dice to be able to spam multiple Mind Blasts several rounds in a row.

[stoutstien]

My personal preference for recharge mechanics for NPCs is having specific triggers that reactive it. Not only does it reduce the cognitive load of having to figure out if something recharges or not or rolling dice to decide it the party can also interact with those triggers to try to avoid them if it's a particularly nasty effect.

[Chronos]

Sure, "recharge on 5-6" is swingy and can result in severe nukeage if the monsters roll lucky, but then, that's also true of "once every 1d4 rounds".

One advantage of 5e's recharge mechanic is that you don't need to keep track of anything. If you're using the 3.x style of "every 1d4 rounds", or a static "ever three rounds" or whatever, then there will come a point where the DM is going "Crap, was that three or four rounds ago that it breath-weaponed last?". With recharge, though, once you've rolled for the round, you can forget about it.

[Skrum]

I really like recharge (usually 5-6), and frequently use it on my custom monsters. I love the tension it creates, getting a roll each round to find out if a boss/powerful monster gets their big gun back. While I haven't asked specifically about it, the players have all reacted very positively to this particular part of the game.

It is so incredibly hard to challenge players in a fun way in 5e. Sure you can just spam save-or-die/suck effects, but that's not fun at all. But something like a breath weapon on a recharge, that the players can then position themselves for to mitigate somewhat, is really fun. Getting an ability back several times, well combat should be hard sometimes.

[Amnestic]

While I can understand not liking monsters being swingy based off of recharge happening or not, there's also the argument that it's good, especially if these monsters aren't just one-offs (eg. mind flayers, in a mind flayer colony). Sometimes they'll be quite deadly, and sometimes they won't be. That sort of variation can help keep players on their toes and helps things be a bit more unpredictable. If mind blast is a 1/SR or whatever, then once they blow it and you save (or fail) you're in the clear, and just have to mop up.

Recharge helps keeps encourage a bit more dynamism - they're neither flush with more resources then they'll ever burn through in 3-4 rounds, nor starved.

I like Recharge, and I wouldn't necessarily be against seeing some features turned into Recharge on player side either.

[KorvinStarmast]

I hate recharge mechanics. I think that recharge mechanics are fit for purpose. The only bit I don't like is that I have to remember to attempt the recharge. It's a bookkeeping thing.

[tKUUNK]

I enjoy the "roll for recharge" mechanic. Have an idea based on this thread though:

Next time I run a monster with 5-6 recharge (or similar) I will roll at the end of the monster's turn, and narratively project the result to the players. This way they'll each have a turn to get positioned, prepared, or heck, try to finish that monster off before its next big attack.

[PhoenixPhyre]

I enjoy the "roll for recharge" mechanic. Have an idea based on this thread though:

Next time I run a monster with 5-6 recharge (or similar) I will roll at the end of the monster's turn, and narratively project the result to the players. This way they'll each have a turn to get positioned, prepared, or heck, try to finish that monster off before its next big attack.

I like this plan. I may adopt it.

[stoutstien]

I enjoy the "roll for recharge" mechanic. Have an idea based on this thread though:

Next time I run a monster with 5-6 recharge (or similar) I will roll at the end of the monster's turn, and narratively project the result to the players. This way they'll each have a turn to get positioned, prepared, or heck, try to finish that monster off before its next big attack.


I like this plan. I may adopt it.

I think this is about the only way you can use a recharge mechanic in a way that is actually interactive.

[rel]

it's simple to run and adds a bit of unpredictable power to combat. I prefer it to limited use per day, which translates to 'spam for NPC's.
Wait mechanics are largely similar and rate about the same.

Other methods I use to limit abilities are come online after x rounds of fighting, activate when bloodied, recharge every other round, etc.

Conditional recharge, e.g. on all attacks missing, on being crit or taking a big single spike of damage, on spending a round out of melee, etc can be interesting.

Another fun way to mix things up is to have a delayed recharge; the ability recharges (visibly) at the end of the creatures turn. So the PC's have a round to prepare before the NPC brings the pain.

And don't be afraid to make your custom powers big and flashy; Have monsters summon mooks to crowd the PC's, explode on death, split apart when killed to reveal new monsters ready to carry on the fight, ground pound to rearrange the battlefield, etc.

Finally, leave the hardcore CC to the PC's. An NPC shouldn't have any abilities that completely deny action.
Slowing or staggering is fine, sleep, paralysis or stunning is not.

[JellyPooga]

I don't like Recharge for the simple reason that using a monster ability a second time in a combat is, well, a bit boring. I much prefer encounters to be dynamic progressions of narrative, rather than static and repetitive, just draining player resources. It's the same with encounter/adventure design; having a goblin den simply be one fight against 5 goblins followed by another and another, it's far better to have a fight against 5 goblins, followed by one against 2 goblins and 3 wolves, followed by a shaman and his bodyguard and so on. The same goes for encounter design. A monster that attacks, then attacks again, then attacks again is obviously pretty dull, but so is a monster that uses its petrifying gaze on one round, then uses it again the next round and again and again. Repetition is boring.

The way I design encounters, or enemies specifically, is based entirely off of how long I think they'll survive in the encounter they're in. If I only expect them to survive 1 round, I'll only give them one, maybe two actions they can use; usually an attack plus something that makes them interesting. If I expect them to last 5 rounds, on the other hand, I'll give them 7, 8 or more actions they can use and within a given encounter, most likely won't repeat any of them unless the PC's really have managed to box them into a corner somehow. It's worth giving a monster more actions than rounds you expect them to live because it gives you wiggle room to use the monster again later and still pull a surprise on the PC's, as well as making them more versatile for multiple scenarios, but it can also include scenario specific actions (drop a portcullis, activate a trap, etc.). If I do feel the need to re-use a specific action or ability, it should be because it's dramatic or significant (e.g. a dragon making a pass over the village breathing fire to burn the buildings and harm the villagers, lending time-pressure to the encounter and/or encouraging the players to engage, then using the same breath attack against the PC's), not because the dice say I can.

The recharge mechanic essentially tells you, the GM, that you should be using the ability again within a single encounter (which again, isn't as interesting as using a different ability), except it also takes away your agency as GM to use it at the dramatically/narratively appropriate time. You can, of course, fudge your rolls or whatever to recharge whenever you want, but the principle of the mechanic doesn't make sense from a gameplay or narrative perspective.

If you have a creature that has an ability, let's use a dragons fire breath as an example, you can showcase the narrative aspect of it without it being the same mechanical feature, making the encounter diverse without just plastering additional whacky features on to your creatures. Breathing a swathe of fire over the party is one action the dragon could take; that's the one we expect and 5e gives us rules for in the MM stat block. Giving the dragon a focused, single target version of their fire breath would be mechanically distinct without having to give them, for example, a second breath type (e.g. ice or repulsion) or funky additional features (e.g. "this dragon has beholder eye-stalks") to make them or the encounter they're in interesting. Giving them another feature that let's them use their breath to burn arrows out of the sky, or create a terrain effect by setting the ground ablaze, or any number of other things you might think of that just that breath weapon could possibly do is far more engaging than simply waiting until you roll a 5-6 on a d6 until you can do that same old AoE fire breath attack.

It's not just fantastical monster abilities either; the same can be applied to regular humanoid creatures too. Goblins that just attack and then attack more (if they're still alive in the following rounds) are boring. Goblins that gang up on a PC, trip them up and dog-pile them into submission are way more interesting. Does it matter that their strength is low? No. They have numbers on their side. Rolling Athletics with advantage (Help action) three or four times in a round is almost guaranteed to grapple and prone a PC (eventually) and is way more interesting than just having them chip away at HP. When one of the PC's is slowly being hauled away by a mob of goblins to a mysterious and dangerous looking pit (who knows what's in there?) and a line of them are blocking the way, it puts pressure on the PC's to do something about it and creates an actual scenario instead of just another boring, HP-depleting fight. Action-activated traps, mundane and magical gear, grapples, shoves, alternate weapons, team-tactics, use of terrain; these are all things that you won't often see in a monster stat-block. They are, however, what will really make an encounter interesting...and to bring it back around to why Recharge is boring...unless you're using them repeatedly within the span of a single encounter. Goblin-piling one PC is fun and interesting. Rinse and repeating it just lacks imagination! :smallwink:

[Atranen]

They've always rubbed me the wrong way, but I haven't put my finger on why. Something feels a bit 'video-gamey' about it, or maybe 'card-gamey', but in any case not what I want out of an RPG. Part of it is that the NPCs and the players are operating on fundamentally different set of rules, which I dislike, because it hurts my verisimilitude. I guess it just feels like *game mechanic* is written across it in big flashing letters, and that takes me out of the game.

How to do it right? Having a certain number of uses per rest (like PCs, good), or having a flat or 1d4 recharge or something both feel better to me. Each has fewer rolls also, which I think alleviates it feeling like a game mechanic.

But in general I prefer monster design that isn't reliant on a single powerful ability with a recharge (which feels like a Legend of Zelda boss). Give dragons a powerful breath weapon, and let them use it a few different ways, and give them access to spellcasting, and you end up with a more interesting and varied opponent.

[PhoenixPhyre]

They've always rubbed me the wrong way, but I haven't put my finger on why. Something feels a bit 'video-gamey' about it, or maybe 'card-gamey', but in any case not what I want out of an RPG. Part of it is that the NPCs and the players are operating on fundamentally different set of rules, which I dislike, because it hurts my verisimilitude. I guess it just feels like *game mechanic* is written across it in big flashing letters, and that takes me out of the game.

How to do it right? Having a certain number of uses per rest (like PCs, good), or having a flat or 1d4 recharge or something both feel better to me. Each has fewer rolls also, which I think alleviates it feeling like a game mechanic.

But in general I prefer monster design that isn't reliant on a single powerful ability with a recharge (which feels like a Legend of Zelda boss). Give dragons a powerful breath weapon, and let them use it a few different ways, and give them access to spellcasting, and you end up with a more interesting and varied opponent.

I'm the other side on both of those. I prefer monsters who have different designs from PCs, because they're just so utterly different in use and balance points. The whole "daily use" thing is already an entirely gamist element.

And IMX, monsters having One Big Thing and then their regular stuff makes them much more likely to do anything interesting. Spellcasting is the worst, because all monsters who are primary spellcasters always blend together, because spellcasting is the least interesting, most bland system ever created. A spellcasting dragon is a spellcaster who happens to be a dragon...or the spellcasting is pointless. And spellcasting makes monsters annoying to run just from a formatting perspective (having to look up stuff in other books).

Monsters have very limited action economy. 1-5 actions total. During their entire on-screen existence. So they should make the most of it, and larding up their stat blocks with stuff they'll never use just makes them harder to run.

[Atranen]

I'm the other side on both of those. I prefer monsters who have different designs from PCs, because they're just so utterly different in use and balance points. The whole "daily use" thing is already an entirely gamist element.

And IMX, monsters having One Big Thing and then their regular stuff makes them much more likely to do anything interesting. Spellcasting is the worst, because all monsters who are primary spellcasters always blend together, because spellcasting is the least interesting, most bland system ever created. A spellcasting dragon is a spellcaster who happens to be a dragon...or the spellcasting is pointless. And spellcasting makes monsters annoying to run just from a formatting perspective (having to look up stuff in other books).

Monsters have very limited action economy. 1-5 actions total. During their entire on-screen existence. So they should make the most of it, and larding up their stat blocks with stuff they'll never use just makes them harder to run.

Hmm, daily use comes from Vancian casting, which was written without "game mechanics" in mind. It's gamist in the sense that everything (dice, hps, AC, whatever) in the game is gamist, but it works narratively in a way that "roll to recharge" doesn't imo.

I get your point in general, but with a big monster like a dragon I'm expecting them to provide more of a challenge than 1-5 rounds of combat. Spellcasting gives them survivability (e.g., to counterspell devastating CC) and out of combat utility that can make them more impactful as villains before they show up on screen. For a guard drake or something the PCs kill on their way to the dragon, One Big Thing is more ok. Imo.

[NichG]

Hmm, daily use comes from Vancian casting, which was written without "game mechanics" in mind. It's gamist in the sense that everything (dice, hps, AC, whatever) in the game is gamist, but it works narratively in a way that "roll to recharge" doesn't imo.


I sort of have the opposite feeling. Daily use makes certain moments of the day special, which maybe you can justify with clerics that have rituals to perform at different specific times, but for arcane casters (not to mention others) its kind of weird. Stuff per long rest is a little better, in the idea that its exhausting somehow and you need to recharge, but things which require separate tracking per long rest suggest the existence of many, many parallel resources existing within someone that in 99% of the population end up un-tapped, but which sequentially end up getting used for this or that specific ability - again kind of odd to me. For both daily and long-rest type resources, a 'pool' system where all of those abilities draw from the same single resource pool ('stamina' or 'life force' or 'mana' or whatever), and resting, eating, etc activities accelerate the refill of that pool feels more narratively natural.

Recharge on the other hand feels very natural to me specifically for things that are opportunistic, or where there's just some kind of bodily timescale to get everything back into position to do it again. For example, its quite natural to experience being briefly winded upon doing something particularly exerting. While that might be a matter of minutes to get over for the average person in real life, its not that hard for me to imagine it would have a different scale for animals, magical beasts, etc. Recharge indicating that the creature (or person) in question has caught their breath, their venom sacs have refilled, they have dumped the residual heat and are ready to act again, their resting level of mental activity has dropped back down below the threshold where they risk a seizure, sure that makes sense to me.

Going a bit less gamist, you could have something like 'Using this power Exerts your X ability score. You cannot use other powers that would Exert the same ability while you are Exerted in such a way, and you suffer disadvantage on ability checks (not saves or attack rolls) using that ability while Exerted. Each round you may attempt a DC 15 saving throw of the same type to remove the Exerted condition.' And honestly that seems like it would work really well for PCs too, for a lot of things that are currently per-short-rest, plus it would give you the extra hooks of being able to potentially design characters around changing their own recharge rates (buff someone's save or burn inspiration or whatever to refresh the big moves faster, etc).

For monsters, you would just design them to have one stat that gets Exerted for ease of tracking.

Alternately, a mechanic that could be interesting for more complex combat maneuvers that rely on positioning and opportunities in a way that isn't resolved would be rather than a recharge, you roll to see which maneuvers are available in a given round, Give each maneuver a number 1..6, and only maneuvers that are within +/-1 of your d6 roll can be used. That would also let you make maneuvers generally more potent.

[PhoenixPhyre]

Hmm, daily use comes from Vancian casting, which was written without "game mechanics" in mind. It's gamist in the sense that everything (dice, hps, AC, whatever) in the game is gamist, but it works narratively in a way that "roll to recharge" doesn't imo.

I get your point in general, but with a big monster like a dragon I'm expecting them to provide more of a challenge than 1-5 rounds of combat. Spellcasting gives them survivability (e.g., to counterspell devastating CC) and out of combat utility that can make them more impactful as villains before they show up on screen. For a guard drake or something the PCs kill on their way to the dragon, One Big Thing is more ok. Imo.

Only one tiny bit of daily use comes from vancian. And vancian makes ~0 sense at all already. For anyone. It's entirely gamist. And always has been--spell slots have always been "artillery ammo."

Even in Vance's work, it wasn't daily at all, it was just simultaneously loaded and took some time to re-learn once spent. But you could "recharge" anytime you had downtime.

I'd expect even a dragon to last only a few rounds, especially if solo. And anything out of combat is (and always has/will be) entirely DM fiat. Doesn't need to be in the blocks at all.

Big "Solos"[1] may have Two Big Things. Any more gets obnoxious to handle. Complexity is best when done compositionally (ie by combining a bunch of simple pieces) rather than by loading down each piece with a bunch of mutually-exclusive options. Because otherwise you end up spending tons of boring DM time trying to figure out what the monster is going to do. At least IMO.

[1] which are crappy fights almost always and always will be, at least with the current way things work. Because action economy is just that important. Yes, even with legendary actions.

[NichG]

[1] which are crappy fights almost always and always will be, at least with the current way things work. Because action economy is just that important. Yes, even with legendary actions.

My current 'ingredients of a solo fight' are that any given solo should have in order to actually work somewhat as an encounter are:

- An always-on passive that interferes with 'standard operating procedure': A battlefield effect or aura or something that changes the rules, either locationally (lava pools, drain your will if you get near, etc) or logically (spells work differently, movement works differently, etc)
- Per round, a single target focus fire action that seriously threatens taking out a single target, either through damage or very severe debuff.
- Per round, an area effect action that either creates a lesser debuff or whittles down the group with damage or both.
- Per round, an ancillary support action that changes the flow of battle somehow - a switchable elemental immunity, summoning, creating turrets.
- Per round, a movement ability of some form.
- Per round, a reactive ability that can be used in response to something someone else does (avoid an attack, counter a spell, curse an attacker, etc)
- Some sort of layered defenses/buffs which can be independently targeted, again either locationally or logically. Three pillars of a ritual which each create a form of DR-based shield, layered magical or environmental defenses that can be dispelled/disjoined - ideally with broad approaches rather than just a specific 'Dispel' spell, etc.
- A once-per-battle 'get out of jail/return to status quo/etc' sort of move. Should be usable reactively. This could be a self-resurrection, a non-action 'remove all status conditions at this point' contingency, a hard stop at 50% health that grants immunity to everything until the solo's next turn, etc.

This list is at least somewhat system-independent.

There's always of course the exercise of 'how did this one being get this sort of range of powers, and why can't each PC do the same thing?'. That depends a lot on setting, but often its some one-off sanctum-like benefits - yes, a PC could set up their sanctum where an entire castle built on a leyline is feeding that into them to get them to take this form. Or it's whatever counts as godhood in that setting, and the methods for getting it go beyond just 'take a level in god' (but if a PC did get a worshipper base, connection to fundamental cosmic concepts, did whatever stuff to ascend then yes they could have this too); or its the effects of some kind of soul-stacking or other multi-creature shenanigans that basically create a joint avatar; or...

[Psyren]

I enjoy the "roll for recharge" mechanic. Have an idea based on this thread though:

Next time I run a monster with 5-6 recharge (or similar) I will roll at the end of the monster's turn, and narratively project the result to the players. This way they'll each have a turn to get positioned, prepared, or heck, try to finish that monster off before its next big attack.


I like this plan. I may adopt it.


I think this is about the only way you can use a recharge mechanic in a way that is actually interactive.

+3! I like this idea.

"The Illithid's eyes glow brighter as the veins in its cranium violently pulse once, almost in warning. You sense it is readying itself for another mental assault on your group."

"As the red dragon roars again, you feel your eyes dry as the dim glow and wave of heat emanating from down its gullet both appear to increase in intensity. It looks like it's starting to inhale..."

[Snowbluff]

I really like recharge (usually 5-6), and frequently use it on my custom monsters. I love the tension it creates, getting a roll each round to find out if a boss/powerful monster gets their big gun back. While I haven't asked specifically about it, the players have all reacted very positively to this particular part of the game.

It is so incredibly hard to challenge players in a fun way in 5e. Sure you can just spam save-or-die/suck effects, but that's not fun at all. But something like a breath weapon on a recharge, that the players can then position themselves for to mitigate somewhat, is really fun. Getting an ability back several times, well combat should be hard sometimes.

Indeed. Also, having some background logic for the monster running definitely helps with making decisions for it. Having a good reason to not do the most brutal or dangerous thing every turn really keeps monsters from being repetitive.

[Atranen]

Recharge on the other hand feels very natural to me specifically for things that are opportunistic, or where there's just some kind of bodily timescale to get everything back into position to do it again. For example, its quite natural to experience being briefly winded upon doing something particularly exerting. While that might be a matter of minutes to get over for the average person in real life, its not that hard for me to imagine it would have a different scale for animals, magical beasts, etc. Recharge indicating that the creature (or person) in question has caught their breath, their venom sacs have refilled, they have dumped the residual heat and are ready to act again, their resting level of mental activity has dropped back down below the threshold where they risk a seizure, sure that makes sense to me.

I get the idea, but suppose a person runs 1 mile at a standard pace, and you then record the amount of time it takes for them to return to their resting heart rate, that time will be standard (for that person)--not at all like a recharge, which happens sometimes in 6 seconds, sometimes in 24. So a flat recharge time works best for me.


I'd expect even a dragon to last only a few rounds, especially if solo. And anything out of combat is (and always has/will be) entirely DM fiat. Doesn't need to be in the blocks at all.

I disagree--if a monster is acting out of combat, I want them to be working according to the mechanics, because this gives the PCs a way to understand what the monster is up to and how to counter it. That means using spell effects as written. I'm ok with variants or additional abilities, but then they should be encoded in the stat block or written down by the DM, and therefore predictable and possible for the PCs to understand.

[BRC]

I enjoy the "roll for recharge" mechanic. Have an idea based on this thread though:

Next time I run a monster with 5-6 recharge (or similar) I will roll at the end of the monster's turn, and narratively project the result to the players. This way they'll each have a turn to get positioned, prepared, or heck, try to finish that monster off before its next big attack.

I like this a lot. I'm always looking for ways to make combats, especially big boss fights, more reactive, giving players things to react to is always good.

[NichG]

I get the idea, but suppose a person runs 1 mile at a standard pace, and you then record the amount of time it takes for them to return to their resting heart rate, that time will be standard (for that person)--not at all like a recharge, which happens sometimes in 6 seconds, sometimes in 24. So a flat recharge time works best for me.

That's not really my experience with exertion personally at least - well at least for being winded, which is different than heart rate. It'll matter how hot or cold it is, how warmly they're dressed, what they ate for breakfast, how much sleep they got, if they were drunk the night before, the degree to which they did warmups and stretches, as well as how recently they've exerted themselves besides. Doing something more complicated than just running, I can easily imagine there being even more unresolved factors. I'd expect quite naturally that when I lift a 200lbs awkwardly-shaped boulder from a certain posture I have a different time to recovery than when I lift a 250lbs object by a handle designed for lifting, which in turn will be different than when I lift and fling an 80lbs goblin at an enemy 5ft away. And if after I exert myself I spend that time just doing mental actions, versus if I'm also swinging a greatsword (vs a rapier, vs a dagger) at enemies around me (who in turn might maneuver through more or less tiring strikes and counter-strikes), well...

Rounds don't give the granularity to resolve things that might be a 1-2 second difference in recovery, it doesn't really do us any good to try to resolve all those details with a list of 10 modifiers and to track all of those little details. So we can just call that 'random variation' since its not like we have the ability to be particularly intentional about it anyhow. It's never going to be worth eschewing your attacks to recharge a quarter of a round earlier for example, because 'a quarter of a round' isn't a thing, so rather than trying to make all of those details part of the decision process, just leave it to dice.

Also just as a larger perspective, measurements having sources of variation is by far the more natural thing than measurements returning exactly the same result every single time. Experiments always have error bars, always, and it takes a lot of work and ensuring perfectly reproduced, clean setups with highly engineered measuring devices to get that variance down.

[Atranen]

Also just as a larger perspective, measurements having sources of variation is by far the more natural thing than measurements returning exactly the same result every single time. Experiments always have error bars, always, and it takes a lot of work and ensuring perfectly reproduced, clean setups with highly engineered measuring devices to get that variance down.

I never said otherwise. But the variance provided by "roll until you get a 5 or 6" is substantially larger than what seems reasonable for recovery time. With that recharge rate, you recover in mean = 3 rounds (median = 1.7) with a standard deviation of ~2.5. That seems overly broad to me.

[PhoenixPhyre]

I'm not disputing this. But the variance provided by "roll until you get a 5 or 6" is substantially larger than what seems reasonable for recovery time. With that recharge rate, you recover in mean = 3 rounds (median = 1.7) with a standard deviation of ~2.5. That seems overly broad to me.

That's 18 seconds, +- 15-ish. When your timescales are that short, wide variance is to be expected.

And saying that everything is fixed is identical to saying that every single dragon, everywhere, has the exact same recharge time on their breath. If it has multiple charges, you're saying that every dragon can breathe exactly X times per day, no more, no less. That's...yeah. Not good worldbuilding. Unless you have a darn good reason why it should be that way.

[Atranen]

That's 18 seconds, +- 15-ish. When your timescales are that short, wide variance is to be expected.

And saying that everything is fixed is identical to saying that every single dragon, everywhere, has the exact same recharge time on their breath. If it has multiple charges, you're saying that every dragon can breathe exactly X times per day, no more, no less. That's...yeah. Not good worldbuilding. Unless you have a darn good reason why it should be that way.

If you sprint 100 yards and see how long it takes to catch your breath, the relative variance will be substantially less. It's not "I exerted myself, and sometimes I can do it again immediately, sometimes it takes quite a while".

With a recharge system, every dragon, everywhere, has exactly the same recharge on their breath. The fact that the distribution is broad does not mean they are pulling from different distributions.

And why can't "Imvaernarhro, master of fire" have a flat recharge of 1 round while other red dragons recharge in 3?

"Red Dragons have an organ that holds a flammable substance which ignites on contact with air. Their bodies produce it slowly and the organ has limited capacity. After breathing a large cone X number of times, the organ will empty and it requires many hours for more of the substance to be produced".

[PhoenixPhyre]

If you sprint 100 yards and see how long it takes to catch your breath, the relative variance will be substantially less. It's not "I exerted myself, and sometimes I can do it again immediately, sometimes it takes quite a while".

With a recharge system, every dragon, everywhere, has exactly the same recharge on their breath. The fact that the distribution is broad does not mean they are pulling from different distributions.

And why can't "Imvaernarhro, master of fire" have a flat recharge of 1 round while other red dragons recharge in 3?

"Red Dragons have an organ that holds a flammable substance which ignites on contact with air. Their bodies produce it slowly and the organ has limited capacity. After breathing a large cone X number of times, the organ will empty and it requires many hours for more of the substance to be produced".

They have the same distribution. But so does everyone who swings a sword, so yeah. Not a big deal. Compared to "you get one and only one, no matter what, full stop, unless you homebrew."

And limited uses per day means you can't have things like
* the dragon uses its breath to destroy a town.
* the dragon strafes the party and retreats, playing hit and run with its superior speed.
* etc.

You drastically make the monster less interesting...and for what? Nothing. Nothing at all.

Edit: and yes, I'd expect wide variation on how long it takes to catch my breath after sprinting 100 yards. Much more than 15 seconds. And that's the point--when the centroid is really low, the relative variance is always high, because the minimum value possible is zero. Half-bounded distributions are not normal when you're shoved up against that bound! It's the absolute variance that matters in this case, not the relative one.

[NichG]

It seems a strange hill to die on compared to other, much larger abstractions. Sure, it could be a bit better to model this with a Gaussian distribution rather than a Poisson distribution, but not so much that its like 'this, this is the thing that breaks my immersion!'. And Gaussian would not be as much of an improvement over Poisson as Poisson is over zero-variance or all of the weirdness of daily charges for something like a breathing fire. The fact that Poisson also gives you a significant reduction in book-keeping compared to the other alternatives because its the one distribution where you don't have to have memory to generate it is a legitimate justification for choosing that abstraction rather than one of the other equally-wrong ones to me. Especially when its used to model something that will happen maybe once per creature ever anyhow...

[Atranen]

They have the same distribution. But so does everyone who swings a sword, so yeah. Not a big deal. Compared to "you get one and only one, no matter what, full stop, unless you homebrew."

And limited uses per day means you can't have things like
* the dragon uses its breath to destroy a town.
* the dragon strafes the party and retreats, playing hit and run with its superior speed.
* etc.

You drastically make the monster less interesting...and for what? Nothing. Nothing at all.

I don't understand why everything having one distribution is better than everything having one value :smallconfused:

Ok. If you want your dragons to use their breath to destroy a town or strafe the party, give them a flat recharge. If you want them to act more like spellcasters, give them X/day. Either is better than variable recharge.


Edit: and yes, I'd expect wide variation on how long it takes to catch my breath after sprinting 100 yards. Much more than 15 seconds. And that's the point--when the centroid is really low, the relative variance is always high, because the minimum value possible is zero. Half-bounded distributions are not normal when you're shoved up against that bound! It's the absolute variance that matters in this case, not the relative one.

I don't think this is true, and you couldn't convince me otherwise without data. Stress tests are used in diagnostics because the response of an individual human is not that variable. Even a simple heart rate recovery rate test has meaningful information. If it was so variable, they'd need to administer these tests several times to the same individual in order to get a meaningful result. But they don't.


It seems a strange hill to die on compared to other, much larger abstractions. Sure, it could be a bit better to model this with a Gaussian distribution rather than a Poisson distribution, but not so much that its like 'this, this is the thing that breaks my immersion!'. And Gaussian would not be as much of an improvement over Poisson as Poisson is over zero-variance or all of the weirdness of daily charges for something like a breathing fire. The fact that Poisson also gives you a significant reduction in book-keeping compared to the other alternatives because its the one distribution where you don't have to have memory to generate it is a legitimate justification for choosing that abstraction rather than one of the other equally-wrong ones to me. Especially when its used to model something that will happen maybe once per creature ever anyhow...

Sure. I disagree with the mechanic for other reasons too :smallbiggrin: it being 'non-physiological' is a small part of it.

[NichG]

I don't understand why everything having one distribution is better than everything having one value :smallconfused:

Because distributions are the natural mathematical object to describe collections of things. The cases where you need to actually model a distribution of distributions exist (e.g. if you have a collection of collections), but 'this set of things all belong to the same distribution' is a tautology for any such set of things. That distribution is called the 'empirical distribution' of that set.

That's different from saying that 'all future things will be drawn from the same distribution' or that a given model in the form of a specific distribution is in fact the generative process leading to those observations. There is some empirical distribution of heights that is 'the distribution of heights of living humans'. That's not to say that there isn't also a distribution of 'the heights of humans who had good nutrition during their upbringing' that would be different, but that exists as a sub-distribution of the broader one. Using the broadest distribution to model the population is just sort of saying 'yes, there can be extra factors, but I acknowledge that I do not know them'.

In that sense, a particular dragon with bad lungs and a dragon with unusual endurance and a dragon who was stabbed 340 years ago by a knight and never fully healed and so on all comprise 'some empirical distribution of dragons', and the minimum intervals between their breath weapon usages would constitute a single distribution. It might be that the dragon with bad lungs should have a longer refresh, the dragon with unusual endurance should have a shorter refresh, etc, but if we say for any given dragon we don't know those details then it only becomes odd to use a single distribution for all dragons once we have a single dragon who is having multiple recharges on screen. Otherwise, we don't have the information as observers to separate variation due to intrinsic differences (epistemic uncertainty) from variation due to the innate stochasticity of the process (aleatoric uncertainty).

If we have sets of dragons, each who recharge multiple times on screen, only then could we actually (in character) be able to infer mathematically that something is odd. That is to say, we could construct a per-dragon model of recharge times and notice that *all* of our uncertainty is aleatoric, which would be peculiar.

Whereas if we use a constant, then we only need to observe multiple dragons recharging just once each to infer that something is odd. The 'oddness' is closer to the surface in that case.

[Atranen]

If we have sets of dragons, each who recharge multiple times on screen, only then could we actually (in character) be able to infer mathematically that something is odd. That is to say, we could construct a per-dragon model of recharge times and notice that *all* of our uncertainty is aleatoric, which would be peculiar.

Whereas if we use a constant, then we only need to observe multiple dragons recharging just once each to infer that something is odd. The 'oddness' is closer to the surface in that case.

Maybe the dragons just had the same lung capacity? That's not particularly odd imo. And it is easy to give them different constants.

[NichG]

Maybe the dragons just had the same lung capacity? That's not particularly odd imo. And it is easy to give them different constants.

It'd be like giving five people IQ tests and having them get exactly identical scores. It could happen, sure, but it starts to become more likely that something weird is going on.

[Derges]

It'd be like giving five people IQ tests and having them get exactly identical scores. It could happen, sure, but it starts to become more likely that something weird is going on.

However, recharge 5-6 would be more like grouping those IQs into 3 buckets. In the case of IQs in <90, 90-110, >110 getting identical results isn't unexpected.

[Xervous]

Guess I’m the weird one who likes the uncertainty of “did the dragon roll a 5 or a 6? Am I feeling lucky?”

[Atranen]

It'd be like giving five people IQ tests and having them get exactly identical scores. It could happen, sure, but it starts to become more likely that something weird is going on.

More like letting two people roll d4s and getting the same value. We're talking about picking from a discrete pool of just a couple numbers. And the scenario was "fighting two dragons makes it odd", not five.

(If it is five, then I raise you "as soon as you fight the same dragon more than once, or even for a long enough span of time that you experience some of the variation, then it becomes clear it is just random and doesn't have a nice explanation like 'dragon Y has good lungs'").

But yeah, this is pretty far in the weeds :smallsmile: you like what you like, I like what I like, and we both (apparently) like to come up with mathematical justifications. And that's fine!

[NichG]

However, recharge 5-6 would be more like grouping those IQs into 3 buckets. In the case of IQs in <90, 90-110, >110 getting identical results isn't unexpected.

Only if you get to observe those people over multiple tests, which is the point I was making before. Because now, the performance on each test will be stochastic, so randomly someone belonging to the lower bin will score into the higher ones.

You can detect the weirdness of a constant recharge fighting five dragons for 5 rounds each, roughly. To detect the weirdness of a constant recharge distribution, you would need to instead fight five dragons 25 rounds each, again roughly.

[crabwizard77]

+3! I like this idea.

"The Illithid's eyes glow brighter as the veins in its cranium violently pulse once, almost in warning. You sense it is readying itself for another mental assault on your group."

"As the red dragon roars again, you feel your eyes dry as the dim glow and wave of heat emanating from down its gullet both appear to increase in intensity. It looks like it's starting to inhale..."

This seems great! I love the telegraph of attacks in videogames, and this seems like a great way to do this in D&D.

[Koury]

As far as the main topic goes, I generally like Recharge. I like it enough that I put it into a fair few abilities in my homebrew Binder class even. As a DM, I like that its one of "those" rolls that the whole table watches (and I'm joining the crowd stealing the whole rolling at then end of turn instead thing). Adds some tension to every round, and some urgency. "Do I clear out some of these annoying mooks, or pile on to the big guy before they absolutely wreck us again?"

[Elenian]

I want an extremely frustrated dragon that hasn't been able to breathe fire for the last five months due to being unable to roll a 5+ (I realize the odds of that happening are essentially zero).

[Darth Credence]

I want an extremely frustrated dragon that hasn't been able to breathe fire for the last five months due to being unable to roll a 5+ (I realize the odds of that happening are essentially zero).

I have a lovelorn blue dragon who can't use his breath weapon any more because "the spark has gone out" since his mate left him. I have weird dragons.

To the topic, I am fine with recharge mechanics, and much prefer to roll them at the start of the turn so we're all surprised by it happening or not. I am not particularly interested in telegraphing the monsters next move a turn in advance, because that would seem to be an incredibly big flaw in fighting ability that creatures that are considered to be powerful combatants just wouldn't have. I think that it makes much more sense for the players to take some risks in regards to it. The players start scattered, so they don't all get hit by the breath weapon, great. But then it's a question of risk for them - if they are better off grouped up, how long do they assume the dragon is recharging for before scattering again? That, to me, is much more dynamic than "we know the dragon can breath again next turn, so lets move apart in preparation". It can also pull out different risk tolerances for the players, as some may be willing to go longer than others before seeking cover.

[RedMage125]

One other method might be "recharges on (event)". 4e used a lot of "recharges on being bloodied for the first time in a fight" (ie dropped to half or less HP)
I know the popular thing is to hate on 4e, but this here is one of the things that made 4e great. That is, Solo monsters, especially dragons. I have run players against dragons in 3e, 4e, and 5e. And 4e were, by far, the most fun. For players and myself.

Dragons not only recharged their breath on getting bloodied, they got an immediate, out of turn use. They had other out of turn actions, as well. It made fights against them feel formidable at any level of play.

[Witty Username]

Predictability is the core of the argument.

Generally per day and fixed recharge are easier to predict, for players and DMs. On the one hand it makes it easier to solve tactical decisions but also to guage the challenge of and work into a larger day.

Variable recharge rates is alot more involved. A dragon that gets its breath weapon every round and another thst only once over the encounter are different enough that they don't really fit the same things. But it is a harder to solve problem.

I think variable recharge as written may be too swingy. A min and max recharge time would be my preference (1d4 round recharge like how 3.5 did it for dragons).

That being said, if the recharge monster is the main stay. Like it is a 5 minute encounter day, recharge as is is fine. Something like a 4-5 level party against a mind flayer where every mind blast is a shock to the system works fine.
Oh winter wolves sound fun, lets use a pack for the medium encounter, into a dozen cold breath attacks in 2 rounds because they rolled smokin. And now the party has to flee the dungeon and long rest for all the damage they weren't expected to take. Is less than ideal.

[Koury]

I think variable recharge as written may be too swingy. A min and max recharge time would be my preference (1d4 round recharge like how 3.5 did it for dragons).

Unless I'm mistaken (which is totally possible!), a 3.5 dragon that rolled 1 round for its breath weapon recovery would be able to breathe again on its very next turn. Similar to how a spell with a 1 round duration would end at the start of your next turn.

Granted, a 3.5 dragon WILL get its breath attack back before too long, where as Recharge 6 could never come up if you're unlucky.

[Amechra]

You can detect the weirdness of a constant recharge fighting five dragons for 5 rounds each, roughly. To detect the weirdness of a constant recharge distribution, you would need to instead fight five dragons 25 rounds each, again roughly.

The thing I'm scratching my head about with regarding this whole part of the conversation is that everyone seems to be ignoring that D&D's combat rounds are pretty chunky. A dragon that recharges after 27-29 seconds and one that recharges after 30-32 seconds would both be represented in-game as dragons that recharge after 5 rounds.


I am not particularly interested in telegraphing the monsters next move a turn in advance, because that would seem to be an incredibly big flaw in fighting ability that creatures that are considered to be powerful combatants just wouldn't have.

As someone who does like the idea... bear in mind that the dragon (or whatever) would be fully aware of how obvious it is that their deadly breath has recovered, and would make plans accordingly. It's a shame that D&D doesn't give you a reason to stick close together like there is in real life, otherwise there'd be an interesting damned-if-you-do-damned-if-you-don't element to the dragon fights.

[Eldariel]

It's a fun PC mechanic. Not really very good for typical monster use, however (a lot to keep track of for relatively little gain; "1d4 rounds" accomplishes the same with 3 times fewer rolls).

[tokek]

It's a fun PC mechanic. Not really very good for typical monster use, however (a lot to keep track of for relatively little gain; "1d4 rounds" accomplishes the same with 3 times fewer rolls).

With the roll d4 you have to keep notes on which round it comes back. The recharge mechanic eliminates some note keeping.

But really the two are very similar and I don’t think many people would have a strong preference between them