The balance issues if we can call them that, not sure that is the correct characterization, but let's call it the progression and state of a game is not dependent on damage and defense as much as action economy.
This is why say an Adult Black Dragon (CR 14) with a budget of 11,500 XP is far less dangerous than say 30 Orcs CR 1/2 with a budget of 3000 XP (x4). 30 Orcs are going to get 30 attacks with a +5 to hit doing 1d12+3. An Adult Dragon is going to get 3 Attacks at +11.
The d20 die is super swingy, you roll that thing 30 times you are going to get 10-15 hits. Thats 10-15d12+3. And that Action Economy is going to be consistent with upswing and downswing. A Dragon might hit all 3 attacks, but again the d20 is swingy, it might just as well miss all 3.
Plus Orcs can lock characters down by sheer overwhelming numbers, preventing fallen characters from getting heals which is critical because one of the problems of 5e is that going to 0 hit points has little impact on the outcome of fights, most characters can heal you and your back in the fight at full strength. A Dragon cant do anything to prevent healing.
There are so many ways action economy is superior to damage and defense. Single monsters in 5e as written are pretty worthless, they simply can't compete against the action economy advantage the players have especially since players often have feats and abilities that allow for bonus actions and special effects.
GMs often think they make monsters stronger by giving them more HP, more damage or more AC, but the only effect this has is that you get the same outcome, it just drags the fight out longer.
You give an Adult Dragon more actions, say one for each player and with each action they can attack 3 times. Now you have a real threat to a party of 4 characters of very high level, 5th level character stand absolutely no chance at all (as they shouldn't) and you don't need to change anything else about the dragon and the fight will be determined much faster.
The balance issues if we can call them that, not sure that is the correct characterization, but let's call it the progression and state of a game is not dependent on damage and defense as much as action economy.
The expectation is that the difficulty of a CR X for a level X party is consistent, and it's not. Two CR 1s vs a level 1 party is a legitimate scary fight. Two CR 20s vs a level 20 party is not (the fact that one CR X is supposed to be a meaningful encounter is a separate math problem that exists at every level).
D&D 5e (can't speak to 5.5e) has one of the highest GM workloads of any given TTRPG. It is literally the worst system to GM. Largely this is due to the high level of player focus in the design.
We've all had those moments where players choose to have their characters do something entirely off the wall. High level play makes that even more likely. To design fun and engaging content post level 12 (in my opinion) is a case of having to fix what the rules designers never did. I'm actually fresh from two 1-19 campaigns that have just finished, and moving onto new 1-20 campaigns. I also run one-shots and shorter runs of a whole load of different systems. Designing encounters that aren't trivial is no small feat with D&D.
I mean even the designers can't get it right. Vecna in Vecna Eve of Ruin, is an absolute pushover. Can be destroyed really easily and really quickly. He's meant to be a big, big bad and yet my players described it as 'anti-climactic doesn't cover it'. Even in that supposed Level 20 end encounter, Vecna is entirely on their own. No minions, no environmental effects, no lair actions, no cool tactical elements. Just 272 hit points of pure boring. If the people paid to look after the IP and design the game can't craft a cool high-level adventure, what chance do you think GMs have?
By contrast, the last Pathfinder game I ran to max level was insane and their Gamemastery Guide was so much better written than the DMG, and a larger array of good 'official' adventures that run through the full array of levels was available.
Heck, even looking at the new Tales of the Valiant Game Master's Guide should have WotC and their staff blushing in absolute shame. That GMG is so well written. Like I was chatting with other GMs as our coupon codes dropped and we literally tried to rip it apart, but just couldn't. I'm seriously looking forward to running a high level game of ToV (with no D&D content).
The problem of high level play is that most GMs who run to high levels like myself, run full campaigns level 1-20. If you didn't get on board with their campaign early you aren't going to get the chance to join in later. Those few GMs who can and do have the skill to run high level one-shots are then also so in demand that they almost don't even have to look for players. In short, the problem is the rules designers and policy people at WotC. They need to actually start considering the DMs who make their game actually possible. I was hopeful that the 2024 DMG might actually help with this problem but after seeing the mess they made of the 2024 PHB - that hope has gone away.
As several people point out here, we don't see many high-level campaigns because they require two things to work: 1) a high level of DM skill to craft and run, and 2) players who employ their characters' incredible powers with restraint. It can be done, however. I just completed playing in a fellow DM's level 11-20 campaign. Fun and challenging.
How did we do it? The game took place in isolated locations, with urgency, that placed limits on resources and downtime activity. Many challenges were best solved through non-combat means, which honestly ran little differently than if we were first level: they required role-play, patience, and creativity. Granted, not every group of players will be on board for this sort of thing. And the DM had a lot of work to prep encounters, usually with large numbers of monsters and homebrew villains to test the mettle of our powerful characters. Believe it or not, we nearly had a TPK at level 20, even with fairly optimized characters and experienced players.
This experience made me want to run a high-level game. It's a thrill to employ 9th level spells or legendary capstone abilities, rather than just read about them.
I am working my way up on a campaign, up to level 13 now, and I think the problem is predictability of the party.
If you know the party already (IE they have worked up to higher levels) then it's going to be much easier to make balanced combats for them to deal with.
EG, if I use a combat calculator and make a series of deadly encounters for the adventuring day for 5 level 18 characters, on the assumption I will pick up players afterwards, so have no idea what characters I'm running for, then it will be impossible for me to know if they are a challenge. Let's say the first combat is a horde of smaller enemies with pack tactics. If the party has lots of magic users with area effects, then dang, that was easy. If they're heavy on rangers, rogues, and barbarians, then dang, this will be a challenge. Next encounter, they are all flying enemies. Welp, fighter and ranger have bows and endurance, barbarian is grumpy, and the spellcasters just did another big spell and made it easier. Now let's pit them against something with resistance to magic, and suddenly a spellcasting party is struggling... do you see the problem? You can't predict if the encounters will actually be difficult or easy, because at high levels there are ways to shut down entire swathes of tactics.
The balance issues if we can call them that, not sure that is the correct characterization, but let's call it the progression and state of a game is not dependent on damage and defense as much as action economy.
The expectation is that the difficulty of a CR X for a level X party is consistent, and it's not. Two CR 1s vs a level 1 party is a legitimate scary fight. Two CR 20s vs a level 20 party is not (the fact that one CR X is supposed to be a meaningful encounter is a separate math problem that exists at every level).
I fully agree, but I still hold that the action economy is the problem. The issue with a 1st level party facing 2 CR 1 monsters is that these have reasonably equal action economies. 20th level characters have a much higher action economy than CR20 monsters. A CR20 monster should be getting 2-4 Actions, potentially 7-10 attacks in a round. Than you have a fair fight. You don't need to change anything else than simply give them more actions in a round.
I have been doing it for years, it completely fixes the CR system. When a CR20 monster has 3 actions and can do 7-10 attacks in a round, 4 20th level characters are going to struggle to survive that fight. Its really that simple to fix.
D&D 5e (can't speak to 5.5e) has one of the highest GM workloads of any given TTRPG. It is literally the worst system to GM. Largely this is due to the high level of player focus in the design.
We've all had those moments where players choose to have their characters do something entirely off the wall. High level play makes that even more likely. To design fun and engaging content post level 12 (in my opinion) is a case of having to fix what the rules designers never did. I'm actually fresh from two 1-19 campaigns that have just finished, and moving onto new 1-20 campaigns. I also run one-shots and shorter runs of a whole load of different systems. Designing encounters that aren't trivial is no small feat with D&D.
I mean even the designers can't get it right. Vecna in Vecna Eve of Ruin, is an absolute pushover. Can be destroyed really easily and really quickly. He's meant to be a big, big bad and yet my players described it as 'anti-climactic doesn't cover it'. Even in that supposed Level 20 end encounter, Vecna is entirely on their own. No minions, no environmental effects, no lair actions, no cool tactical elements. Just 272 hit points of pure boring. If the people paid to look after the IP and design the game can't craft a cool high-level adventure, what chance do you think GMs have?
This is true but only if you are trying to resolve balance exclusively by the rules. Meaning that if you are building encounters with the intent to run them as written, balance especially at higher levels is unlikely and even at lower levels its mostly coincidental.
Adjusting encounters to suit the party is just a part of the game, but its not an exclusive "thing" to D&D, this is a problem in most games.
The issue is that the advice for GMing high-level encounters is terrible. There is this consensus of upping AC, Hit Points and Damage and these things do not solve most problems because they are not the cause, they just drag out fights. A D&D battle should last 3 rounds, no more and those 3 rounds should be filled with nail-biting dice rolls. Give Vecna 3-4 actions per round. That's all you have to do to turn that encounter into a serious problem for any max-level adventuring party.
The reason this works is because downing a 20th level character in a D&D encounter is completetly meaningless. They will be back on their feat and taking actions the very next round and players can bounce of like this repeadly. A deadly encounter should be just that, a 50-50 in which players should have a clear expectation that some characters will die, their might be a TPK and the odds on their success are a crap shoot at best.
The meaning of deadly is that people will die everytime. If that doesn't happen, its not a deadly encounter.
The reason this works is because downing a 20th level character in a D&D encounter is completetly meaningless. They will be back on their feat and taking actions the very next round and players can bounce of like this repeadly. A deadly encounter should be just that, a 50-50 in which players should have a clear expectation that some characters will die, their might be a TPK and the odds on their success are a crap shoot at best.
This is, to my mind, a highlight of a separate issue - the pop-up combat mechanics. This is something I have worked to solve with my Gritty, Consequential and Impactful Dying Rules, which make it much more beneficial to heal people whilst they are alive than when they have already fallen! It also includes lingering injuries which, again, make people want to stay upright, rather than the current mentality of "don't heal me yet, wait until I die, it'll be more useful then!".
The problem as-is is that healing someone for 20hp when they are at 4hp, and them then taking 25hp and dying, is less efficient than letting them take 25hp whilst at 4hp, die, and then pop them up with 20hp of healing.
I do like the idea of the additional actions, but isn't this what Legendary actions do already? Plus lair actions for bosses on home-turf?
When I'm homebrewing I like to include the bosses main abilities as legendary actions. For example, I made the Doom Slayer for a oneshot, and one of his legendary actions was to charge the BFG. He couldn't fire it unless he charged it, so it was telegraphed in a legendary action. I also like to have at least one Reaction and one Bonus Action on each monster when I write them - too many have an action and that's it, with multiattack, making other actions worthless in combat.
one of the problems of 5e is that going to 0 hit points has little impact on the outcome of fights, most characters can heal you and your back in the fight at full strength.
Again that is not a problem, it is a feature. D&D 5e is not a challenging tactical combat game, it is a "hey remember that time we killed a dragon by dropping an anvil on it's head?"-game. It is designed for a group of friends getting together, eating snacks, drinking, and just chilling and having a good time. It is more akin to Skyrim than to Elden Ring. It deliberately doesn't require skill, smarts or tactical genius to play. It deliberately is almost impossible to die / TPK. It is not meant to be challenging, it is not meant to have a significant risk of failure, it is not meant to require teamwork and coordination. It is a power fantasy where players can make whatever character they want and have that character become a monster-killing superhero. It is meant to be able to be played by groups where nobody has a thorough understanding of the rules, and everyone's a little tipsy or on a sugar rush, or the whole party are 10 year olds.
CR is designed so that a party of randomly assembled 10 year olds with whacky characters designed to emulate their favourite superheroes can fight & kill a dragon after school. Or so a group of first-time gamers can get together after work at a pub, and play as their fantasy ideal versions of themselves. Or so that guy who says they want to play the "worst character" can still be an effective member of the party, get to occasionally be the hero and have a good time.
If the CR system was designed such that it lead to almost-TPKs even among experienced players with optimized characters with Legendary magic items, then it wouldn't work for casual players who just want to muck around and try crazy plans like throwing a vampire into the River Styx. See it is impossible to please everyone? So the game designers wisely choose to make the guide books designed for those casual players who don't know what they are doing so don't know how to bend / change the rules without breaking the game or who aren't going to be spending 4-5 hours prepping for a game, and left the serious tactical players to figure out for themselves how to make tactical challenging combats that suit their tasted. Because those first-time gamers or tipsy frat boys are going to need to refer to the books a lot more than the table of tactical geniuses who have spend hours studying the rules to make the most powerful character they can.
If the CR system was designed such that it lead to almost-TPKs even among experienced players with optimized characters with Legendary magic items, then it wouldn't work for casual players who just want to muck around and try crazy plans like throwing a vampire into the River Styx. See it is impossible to please everyone? So the game designers wisely choose to make the guide books designed for those casual players who don't know what they are doing so don't know how to bend / change the rules without breaking the game or who aren't going to be spending 4-5 hours prepping for a game, and left the serious tactical players to figure out for themselves how to make tactical challenging combats that suit their tasted. Because those first-time gamers or tipsy frat boys are going to need to refer to the books a lot more than the table of tactical geniuses who have spend hours studying the rules to make the most powerful character they can.
So, this is where I once again point to Wave Echo Cave as an example of the game as designed. That single dungeon is a great model of the game working and planned as intended. The lesser monsters throughout absolutely bleed away character resources. This is the reason that Neznar, the black spider, is so underwhelming. The desingers understood that the party were likely extremely low on resources by the time the reached that final boss. Assuming too, that the party only short rest twice in their run through those caves it can actually be a rewarding fight. It's never going to be an easy TPK for the black spider, but it will be a tough one for the party. That's the game working as designed. For years, I know I never ran that dungeon as intended with characters taking short rests whenever they felt like. It was only when limiting them to two short rests per in game day that Wave Echo Cave finally made sense to me.
The same frankly is true of high level play. There are underlying assumptions that the designers have made. Mostly they are carry overs from previous versions of the game. The long dungeon crawl is actually core to the assumptions made in the design of D&D 5e. The final villain's lair ought to be jam packed with enemies. It ought to be unsafe to take a long rest in. There is however, by any measure, a player entitlement issue with D&D specifically. And the designers have continually played into it. That's not a bad thing necessarily let's be clear. Giving players more options and more choices is generally a good thing. It is the implementation that becomes an issue.
Let's put together a hypothetical adventuring day for high level though as a thought experiment. We're assuming 5 adventurers here, all at level 19. Daily XP budget is 150,000 and we need to get 6 medium to hard encounters within that budget. This will be a dungeon that is undead themed.
This is where we can see the designers messing up the DMG. They didn't consider that at high levels a larger daily XP budget was going to be needed. We're effectively likely to go over our budget here with our final of the 6 encounters today. However, we can also see that the assumptions of the DMG are based on a party of 6, not of 5. (For me, 5 players is typical which is why I'm showing off a day for 5 players). So our XP numbers are slightly incorrect.
This is all medium difficulty encounters. No hard encounters here, all medium encounter difficulty. Now, I've actually run this adventuring day, and I did so with a Barb, Wizard, Rogue, Cleric, Paladin. No multiclass, no feats, and one magic item each. Even this adventuring day was fairly easy and fairly survivable for the party. In fact there is a massive gulf between this rather easy adventuring day, and an adventuring day that would lead to a TPK. This adventuring day ended with no-one having dropped unconscious, and only the Barbarian having lower than max hit points! See what I mean about easy?
Fact is that the rules designers just weren't great at designing the fundamentals of the game at the higher end. My assessment of such is that they have given too many shiny things to players, but not counterbalanced with cool monster effects. So, you're kinda left with only one choice as a GM - more, and difficult encounters designed to wick away resources to the point that they're almost tapped out by the time the reach the final boss. It's effectively a common limitation. By the by, I'd consider simply throwing more monsters along with the end boss to be part of this.
Simple answer to why so few people run high level though is because Crawford et. al suck at their jobs. The DMG is garbage.
Again that is not a problem, it is a feature. D&D 5e is not a challenging tactical combat game, it is a "hey remember that time we killed a dragon by dropping an anvil on it's head?"-game.
While it's absolutely possible to have a beer and pretzels D&D game, it's really not the way the game is designed -- if it was, we certainly wouldn't have concepts like the adventuring day (I assure you, tracking multiple different resources that can be depleted across multiple encounters and probably multiple sessions is not what casual gamers want to do) and high levels would be much simpler (casual players also don't want to shuffle through their 20+ prepared spells and 100+ available spells).
If you're not doing major reworks to the game system, that should mean that the encounter system should be designed to be lenient at the levels normally played by casual gamer (i.e. low level) and increasingly less forgiving as levels increase (as, presumably, people who play at higher levels are comfortable with higher difficulty).
However, we can also see that the assumptions of the DMG are based on a party of 6, not of 5.
Citation needed. When I looked through the DMG all I could find are:
"Party Size: The preceding guidelines assume that you have a party consisting of three to five adventurers" - So your party is at the upper end of power due to the large number of players,
Note that:
"Increase the difficulty of the encounter by one step (from easy to medium, for example) if the characters have a drawback that their enemies don’t. Reduce the difficulty by one step if the characters have a benefit that their enemies don’t." - This includes surprise, so if your party managed to get surprise your encounters were not "Medium" difficulty they were Easy.
Also note that:
"This adjusted value is not what the monsters are worth in terms of XP; the adjusted value’s only purpose is to help you accurately assess the encounter’s difficulty." and: "Use the Adventuring Day XP table to estimate how much XP that character is expected to earn in a day."
Your example Adventuring day only earns the party:110,300 XP / 150,000 XP Budget. -> i.e. an Easy-ish adventuring day which is exactly what you observed.
"This adjusted value is not what the monsters are worth in terms of XP; the adjusted value’s only purpose is to help you accurately assess the encounter’s difficulty." and: "Use the Adventuring Day XP table to estimate how much XP that character is expected to earn in a day."
Two sentences later we see "This provides a rough estimate of the adjusted XP value for encounters the party can handle before the characters will need to take a long rest." So no, he's doing it correctly.
Let's put together a hypothetical adventuring day for high level though as a thought experiment. We're assuming 5 adventurers here, all at level 19. Daily XP budget is 150,000 and we need to get 6 medium to hard encounters within that budget. This will be a dungeon that is undead themed.
OK! But this time let's do it properly.
Total Daily XP budget for monsters = 150,000 XP Using the same monsters as you this was my adjustment to get the appropriate total: Boneclaw = 8400 XP * 2 = 25200 Devourer = 10000 XP = 10000 Sword Wraith Commander = 3900 XP = 3900 Sword Wraith Warrior = 700 XP *5 = 3500 Revenant = 1800 XP * 2 = 3600 Beholder Zombie = 1800 XP * 5 = 9000 Deathlock Mastermind = 3900 XP = 3900 Death Tyrant = 11500 *3 = 34500 Nightwalker = 25000 = 35000 Demilich = 20000 * 2 = 40000 Total: 150,200 XP
Now to assemble them into encounters: Encounter 1:Sword Wraith Commander, 5x Sword Wraith Warrior, 2x Revenant = 22,000 = Medium encounter Encounter 2: 2x Bone Claw + Devourer = 39,300 XP = Hard encounter Short Rest 1 happens here, as it's about 1/3rd of the way through the adventuring day.
Encounter 3: 5x Beholder Zombie, Deathlock Mastermind = 25,800 = Medium encounter Encounter 4: 3x Death Tyrant = 69,000 = Deadly encounter Short Rest 2 happens here, as we assume we're now 2/3rds of the adventuring day done
Encounter 5: Nightwalker = 25,000 = Medium encounter Encounter 6: 2x Demilich = 60,000 XP = Deadly encounter
I'd be super curious what the results would be if you reran the test with this appropriate adventuring day based on the DMG rules. But again make sure they don't have any magic items as magic weapons negate one of the Demilich's (and some of the other monster's) resistances which are a core part of its defenses. - Though honestly the Demilich is pretty terrible, you need to have that encounter in a tiny room or hallway in a tomb, I'd replace it with a Lich + 3* Bodaks if playing in a large arena.
I'd be super curious what the results would be if you reran the test with this appropriate adventuring day based on the DMG rules.
It won't change, because he used the rules correctly. Also, using your interpretation, bear in mind that by your logic this is an accurate 1st level adventuring day for 5 PCs (daily budget: 300x5 = 1500)
8x guard (25 xp each, xp 200, adjusted xp 500) (deadly)
8x stirge (25 xp each, xp 200, adjusted xp 500)
8x kobold (25 xp each, xp 200, adjusted xp 500) (at this point, we've reached the daily limit based on adjusted xp)
I had a DM in my epic game start us 300ft from the encounter the valley was an anti teleport zone so no DD no teleport no misty step etc. We had to use terrain for cover when possible or take huge ballista attacks off the castle walls before the fight even started (L25 game)
Or you just walk a little ways away and the Druid uses Wind Walk and Pass without Trace and you fly past the combat up in the sky effectively hiding amongst the clouds and sneak down to change back in a dark corner of the castle hidden by Pass without Trace. Or you all step back, they cast Conjure Animals to give you all mounts with 80ft movement speeds that can Dash every round to get you there in 2 turns. Or they use Animal Shapes to turn you all into inconspicuous mice and you just run up to the walls and through a sewer pipe into the castle.
Just a couple of comments since I have nothing to do with the OPs game. However, in this case it is a level 25 encounter. What makes you think that any of these feeble ideas would have any reasonable chance of success?
1) Wind walk + pass without trace. Wind Walk doesn't make you invisible. Those oddly moving clouds in the sky are likely to stand out no matter what else is happening especially if the castle is alert. Pass without Trace does NOTHING if you are visible. So ... they will get noticed and wind walk doesn't make them invulnerable so they get to be targeted by whatever defenses can aim towards the sky. In addition, the players were unaware the castle is actually a sentient creature that is likely to notice ANYTHING entering.
2) Mounts reduce your time on the field and get you closer faster but don't really do anything else. They also die quickly. Also by that level there are likely characters with base movement and magic that are just as fast. Running in quickly is likely a non-issue and good luck keeping out an eye for traps dashing towards the castle at 160'/rd. I'm assuming the kind of traps found at L25 are likely nasty enough that avoiding them would be preferable to running over them.
3) Inconspicuous mice that can be seen by detect magic. Which mundane pets can see and attack ... trained raptors in the air. Keep in mind that the sewer pipe you are talking about is a part of the sentient castle. Interesting when the mice are swallowed and confined.
Or the Wizard/Sorcerer/Bard/Warlock turns the whole party invisible, you walk through the valley undetected and sneak in a back door. (bonus points for Fly + Invisibility on the whole party if you have two such casters so you don't even need to use a door). Or they use Seeming to disguise all of you as inconspicuous travellers / merchants / soldiers of the castle and you walk up without getting attacked.
4) True Seeing is common at that level. See Invisible itself is only a level 2 spell. Against opponents of this nature one would expect most of the guards to be able to see invisible and I would expect the castle itself will have both true seeing and blindsight.
5) Seeming is an illusion and is bypassed by True Seeing which at least some of the guards would be expected to have. Minimum.
It seems possible that you've never played or run a game for high level character and the shenanigans that they can get up to. My home game is currently level 18 and the Adventurer's League run of Out of the Abyss I am running is at 12th. I've also played a fair bit at tier 3 and a few tier 4s - so I've seen what can be done and also what a DM will have in mind that counter's all the suggestions you made above unless the party does something more than that.
Or you just have a sniper duel with long range spells and ranged attacks: fall prone to give the ballista DA behind a rock so you have 3/4 cover, then stand up, fire EB with the extended range invocation/feat or a longbow with SS, or spells like ice storm.
6) This assumes that the enemy gives you a target and there isn't some reason you actually need to get close ... like rescue a princess?
It's not really "high level play" if the DM just arbitrarily bans all the high level abilities that trivialize the encounter. That's just playing low level D&D with bigger numbers.
The same can be said for: "mind you this is after a lot of heavy duty spells where blown on the 2 mini bosses that where resistant to basically everything in the courtyard", ooh wow you deal twice as much damage as a 10th level character so now the bosses just take 1/2 damage and deal 2x as much damage as their 10th level counter parts - that's just Red-Queening. Combat is not more exciting if the enemies deal 100 damage vs your 1000 hit point than if they deal 10 damage vs your 100 hit points or 1 damage vs 10 hit points.
The DM doesn't need to ban anything. The players need to come up with different plans since the low level tactics require modifications to be successful. Mind blank, the spell to mask magical signatures, perhaps finding a herb to mask your scents - LOTS of things to do but the simple cast a spell and solve the problem is usually off the table.
The group thoroughly enjoyed the encounter and at the end of the session when we thought we won the castle came alive and we learned it was a sentient structure that was now going to try and kills us. So next session we are starting at below 50% spells slots and trying to kill a sentient castle to save a princess.
Don't you find that immersion breaking? If the castle was sentient why did it wait to do anything until after you killed all the other creatures inside of it? Why would the castle favour the current occupants over your party? Rather than trying to kill the castle your party should just talk to it if it is sentient and convince it to leave you alone, at level 25 you should easily be able to succeed on a DC 30 Persuasion check. Or just cast Charm Monster or Suggestion on it.
They "thought they had won" ... this doesn't necessarily mean they marched through and cleaned out the castle ... it may have meant they silenced the wall defenses and thought they just needed to enter the castle and mop up. Ooops ... DM had additional twists in mind. My guess is that the sentient castle is just the next line of defence after the walls and there are likely more. I I was running it, I'd guess that the princess is on a Demiplane and that if they get through the castle they can find a portal or some method of access to the next part of the adventure ...
Also doesn't seem like raiding that entire castle was actually a challenge despite you guys seeming to do it the hardest possible way if you're only down 50% of your spell slots.
I'd suggest that there may have been other methods possible but none of the ones you suggested were at all likely to work in a level 20+ game.
However, we can also see that the assumptions of the DMG are based on a party of 6, not of 5.
Citation needed. When I looked through the DMG all I could find are:
"Party Size: The preceding guidelines assume that you have a party consisting of three to five adventurers" - So your party is at the upper end of power due to the large number of players,
Note that:
"Increase the difficulty of the encounter by one step (from easy to medium, for example) if the characters have a drawback that their enemies don’t. Reduce the difficulty by one step if the characters have a benefit that their enemies don’t." - This includes surprise, so if your party managed to get surprise your encounters were not "Medium" difficulty they were Easy.
Also note that:
"This adjusted value is not what the monsters are worth in terms of XP; the adjusted value’s only purpose is to help you accurately assess the encounter’s difficulty." and: "Use the Adventuring Day XP table to estimate how much XP that character is expected to earn in a day."
Your example Adventuring day only earns the party:110,300 XP / 150,000 XP Budget. -> i.e. an Easy-ish adventuring day which is exactly what you observed.
On the subject of your 'proper' encounter You have added in Deadly encounters which automatically skew the number of encounters we should have. Pages 81-84 of the DMG (first Printing, December 2014) are where to look for this information. If you're throwing in 'deadly' encounters it means you need fewer encounters during the adventuring day. If you follow the assumptions laid out. Now, without actually running the encounter, I'm fairly sure that even what you've proposed is pretty easily survivable.
At best guess, even assuming the enemies always get surprise, the party survive easily. Now, I will actually run this adventuring day with my group of GMs (we playtest stuff quite regularly these days), but I guarantee that none of these encounters will run for more than 6 rounds. For the Demilich, consider that the barb might well have functions that turn their attacks into magical attacks. Paladin's smite will cleave through some of those immunities and resistances too. We have both Cleric and Paladin represented here which gives some benefits against undeaed too. In this instance I've picked a quite difficult, if basic team of adventurers for enemies to go up against. I would even go so far as to claim that even with a team of just four - this adventure day is totally survivable. Probably pretty easy too. Based off experience, even losing the Paladin from the equations, the party aren't going to have a single member fall unconscious.
I did use the rules correctly, and would suggest taking a look again at the Encounter section of the DMG.
However you work it play at tiers 3+ can be difficult for a few reasons - 1) characters are extremely capable and players are generally experienced to highly experienced so they are capable of lots of out of the box solutions you may not be prepared for - no matter how prepared you are. 2) the CR system tends to be out of whack at high level so it takes more DM work to make it work - and DMs don’t all want to do or have time to do that much work on top of any/everything else they are already doing. 3) making the sorts of adjustments as well as prepping for the shenanigans of the players takes experience to do well so everyone enjoys it. Most DMs have under 10 years experience (many less than 5) mostly at lower levels of play so they tend to be unprepared for this. Only 3.x and 4e made any efforts at higher level play. 3e had the epic players guide (essential for post L20 play and a decent starting point for high level play. 4e allowed PCs to progress to L30 which covered the vast majority of tier 3+ play. Even in those high level play was problematic for many of the reasons given above. To do tier 3+ campaigns basically calls for experienced players and highly experienced DMs and while there are lots numerically given the number of folks playing there aren’t that many in most locations so you don’t see that many games. Even for those that are most of our games are at lower levels as those can be just as fun so why go to the effort?
I had a DM in my epic game start us 300ft from the encounter the valley was an anti teleport zone so no DD no teleport no misty step etc. We had to use terrain for cover when possible or take huge ballista attacks off the castle walls before the fight even started (L25 game)
Or you just walk a little ways away and the Druid uses Wind Walk and Pass without Trace and you fly past the combat up in the sky effectively hiding amongst the clouds and sneak down to change back in a dark corner of the castle hidden by Pass without Trace. Or you all step back, they cast Conjure Animals to give you all mounts with 80ft movement speeds that can Dash every round to get you there in 2 turns. Or they use Animal Shapes to turn you all into inconspicuous mice and you just run up to the walls and through a sewer pipe into the castle.
Just a couple of comments since I have nothing to do with the OPs game. However, in this case it is a level 25 encounter. What makes you think that any of these feeble ideas would have any reasonable chance of success?
1) Wind walk + pass without trace. Wind Walk doesn't make you invisible. Those oddly moving clouds in the sky are likely to stand out no matter what else is happening especially if the castle is alert. Pass without Trace does NOTHING if you are visible. So ... they will get noticed and wind walk doesn't make them invulnerable so they get to be targeted by whatever defenses can aim towards the sky. In addition, the players were unaware the castle is actually a sentient creature that is likely to notice ANYTHING entering.
2) Mounts reduce your time on the field and get you closer faster but don't really do anything else. They also die quickly. Also by that level there are likely characters with base movement and magic that are just as fast. Running in quickly is likely a non-issue and good luck keeping out an eye for traps dashing towards the castle at 160'/rd. I'm assuming the kind of traps found at L25 are likely nasty enough that avoiding them would be preferable to running over them.
3) Inconspicuous mice that can be seen by detect magic. Which mundane pets can see and attack ... trained raptors in the air. Keep in mind that the sewer pipe you are talking about is a part of the sentient castle. Interesting when the mice are swallowed and confined.
Or the Wizard/Sorcerer/Bard/Warlock turns the whole party invisible, you walk through the valley undetected and sneak in a back door. (bonus points for Fly + Invisibility on the whole party if you have two such casters so you don't even need to use a door). Or they use Seeming to disguise all of you as inconspicuous travellers / merchants / soldiers of the castle and you walk up without getting attacked.
4) True Seeing is common at that level. See Invisible itself is only a level 2 spell. Against opponents of this nature one would expect most of the guards to be able to see invisible and I would expect the castle itself will have both true seeing and blindsight.
5) Seeming is an illusion and is bypassed by True Seeing which at least some of the guards would be expected to have. Minimum.
It seems possible that you've never played or run a game for high level character and the shenanigans that they can get up to. My home game is currently level 18 and the Adventurer's League run of Out of the Abyss I am running is at 12th. I've also played a fair bit at tier 3 and a few tier 4s - so I've seen what can be done and also what a DM will have in mind that counter's all the suggestions you made above unless the party does something more than that.
Or you just have a sniper duel with long range spells and ranged attacks: fall prone to give the ballista DA behind a rock so you have 3/4 cover, then stand up, fire EB with the extended range invocation/feat or a longbow with SS, or spells like ice storm.
6) This assumes that the enemy gives you a target and there isn't some reason you actually need to get close ... like rescue a princess?
It's not really "high level play" if the DM just arbitrarily bans all the high level abilities that trivialize the encounter. That's just playing low level D&D with bigger numbers.
The same can be said for: "mind you this is after a lot of heavy duty spells where blown on the 2 mini bosses that where resistant to basically everything in the courtyard", ooh wow you deal twice as much damage as a 10th level character so now the bosses just take 1/2 damage and deal 2x as much damage as their 10th level counter parts - that's just Red-Queening. Combat is not more exciting if the enemies deal 100 damage vs your 1000 hit point than if they deal 10 damage vs your 100 hit points or 1 damage vs 10 hit points.
The DM doesn't need to ban anything. The players need to come up with different plans since the low level tactics require modifications to be successful. Mind blank, the spell to mask magical signatures, perhaps finding a herb to mask your scents - LOTS of things to do but the simple cast a spell and solve the problem is usually off the table.
The group thoroughly enjoyed the encounter and at the end of the session when we thought we won the castle came alive and we learned it was a sentient structure that was now going to try and kills us. So next session we are starting at below 50% spells slots and trying to kill a sentient castle to save a princess.
Don't you find that immersion breaking? If the castle was sentient why did it wait to do anything until after you killed all the other creatures inside of it? Why would the castle favour the current occupants over your party? Rather than trying to kill the castle your party should just talk to it if it is sentient and convince it to leave you alone, at level 25 you should easily be able to succeed on a DC 30 Persuasion check. Or just cast Charm Monster or Suggestion on it.
They "thought they had won" ... this doesn't necessarily mean they marched through and cleaned out the castle ... it may have meant they silenced the wall defenses and thought they just needed to enter the castle and mop up. Ooops ... DM had additional twists in mind. My guess is that the sentient castle is just the next line of defence after the walls and there are likely more. I I was running it, I'd guess that the princess is on a Demiplane and that if they get through the castle they can find a portal or some method of access to the next part of the adventure ...
Also doesn't seem like raiding that entire castle was actually a challenge despite you guys seeming to do it the hardest possible way if you're only down 50% of your spell slots.
I'd suggest that there may have been other methods possible but none of the ones you suggested were at all likely to work in a level 20+ game.
David42 is absolutely correct a few points for clarification then I will let this go as the notion that this stuff did not occur to a group of seasoned gamers is incorrect.
A very high percentage of random mobs and every mini boss or boss in a l25 game should be assumed to have most if not all the following - Alert, True sight, Multi attack, Cast + Melee same round with various attacks, no disadvantage for casting within 5', Fly on demand, Tier 1 and Tier 2 Epic Spells, various resistances. Passive Insight, Perception and Investigation 25+, to hit and save modifiers of 12+, outright immunity to 1 damage type. Potentially reflective damage, potentially healed by one type of damage.
There could have been traps, a horde of invisible mobs lying in ambush between us and the castle, or anything that the dm could have dreamed up - proceeding cautiously is always best. We cleared the field - the sorcerer animated some boulders and had them go in front of us as cover- we took some damage on the way in but where at least reasonably certain there was nothing behind us then proceeded to deal with the guards, ballista's, mini bosses etc.
In this case these mini bosses had broad resistances - some times its elemental only - sometimes its force - sometimes its whatever random combination of resistances ----as a group the key is to figure out what they are resistant to then leverage that. See what saves they are bad at - then leverage that etc etc its not a low level game with big numbers you literally have the entire PHB DMG and Epic ruleset at your disposal the level of creativity possible is staggering.
Plane hopping - world hopping - continent hopping all happen on a regular basis my guess is that when we kill the castle the force that animated it will be dispelled and the princess will be inside - then again maybe not the DM made it up so its his story to tell. Likely the evil force that animated the castle is a minion of the ultimate threat and the next task will be to go hunt that down.
As an aside much has been made of the CR system and action efficiency - in most high level games it is assumed that the fight difficulty is adjusted by the DM prior to each session given the groups overall strength--- a combination of AC - spike damage - continuous damage - legendary skills - lair skills - high dc saves - and unique homebrew skills and spells all blend together to make a fascinating adventure.
This is a very basic example but illustrative of my point --- you kill a group of 4 seemingly normal fire elementals as soon as the 4th dies the embers from all 4 reform in to 2 xl fire elementals kill those 2 then the embers reform into an xxl elemental - its a common tactic and makes parties be cautious about not dropping everything on 1 part of a fight. Assuming that it is all correctly balanced it makes for a thrilling fight that the party has a reasonable chance of winning.
There has been a lot of great discussion in this thread I will just add that I hope more folks get motivated to play in and run these games.
Right now if somehow my level 5 party would win initiative against an adult white dragon, the dragon would be dead in two rounds. That isn't factoring in the New Weapon Masteries. If the players can position themselves where the breath weapon only takes one or two of them out, they are good. But the breath weapon can do anywhere from 12-96. Its breath weapon is its only attack that would scare my level 5 party. They will kill the dragon long before it has any chance of killing them. To limit the breath weapon you only need to be circling the dragon up close. Weapon Masteries could prevent the dragon from flying away.
So a CR 13 creature can be killed by level 5 PCs with almost no magic items. (They have like a +1 weapon). While in an actual game, its unlikely all the PCs will go before the dragon. I think it still illustrates the problem, that PCs quickly become more powerful than the monsters they are going to be facing. Which means, more monsters are needed. at Level 11 or 12, I had an encounter go all 4 hours of the game session. So with only 4 hours to play, if we did the 6-8 encounters per day, we would be looking at probably an entire month of play that is still just combat encounters.
The game is designed for Dungeon Crawls, and selling books. So player options become more and more powerful, which the new 5e makes even worse. and the solution is to have more and more encounters to remove player resources. which works great for dungeon crawls. Doesn't work so great when you are taking on a singular dragon or in a location that really can't or shouldn't have 8 encounters.
5e has a monster design problem, mainly with how the action economy works among other things. This problem was identified a long time ago, but this edition of the game is going on 10 years with little response from WotC on it. I hope Wizards of the Coast has learned a thing or two about monster design in that time and produces a better Monster book, but the good news is that there is MCDM's Flee Mortals.
This book addresses the three key design issues with the official D&D monster manual.
* Singular monster action economy problems * Boring bags of hit points problem * Key word reference problem
It does this by creating action-oriented monsters, eliminating keywords and giving you executable powers right with the stat block and creating diverse versions of monsters so you always have several Goblins, Orcs... etc.. to pick from. oh and minions, thank god for minions!
I can honestly say that this book, saved 5e D&D. It makes fighting monsters fun again.
The monsters are tough as hell too. A CR 13 Dragon in Fleet Mortals you might be able to defeat if you have a party of 4 13th level characters... maybe... if they are hyper-optimized and play super tactical. A 5th level party might not even get to act before they are all killed even if they win the initiative.
I think Flee Mortals is better, but I have still had issues with those monsters not being a threat.
The balance issues if we can call them that, not sure that is the correct characterization, but let's call it the progression and state of a game is not dependent on damage and defense as much as action economy.
This is why say an Adult Black Dragon (CR 14) with a budget of 11,500 XP is far less dangerous than say 30 Orcs CR 1/2 with a budget of 3000 XP (x4). 30 Orcs are going to get 30 attacks with a +5 to hit doing 1d12+3. An Adult Dragon is going to get 3 Attacks at +11.
The d20 die is super swingy, you roll that thing 30 times you are going to get 10-15 hits. Thats 10-15d12+3. And that Action Economy is going to be consistent with upswing and downswing. A Dragon might hit all 3 attacks, but again the d20 is swingy, it might just as well miss all 3.
Plus Orcs can lock characters down by sheer overwhelming numbers, preventing fallen characters from getting heals which is critical because one of the problems of 5e is that going to 0 hit points has little impact on the outcome of fights, most characters can heal you and your back in the fight at full strength. A Dragon cant do anything to prevent healing.
There are so many ways action economy is superior to damage and defense. Single monsters in 5e as written are pretty worthless, they simply can't compete against the action economy advantage the players have especially since players often have feats and abilities that allow for bonus actions and special effects.
GMs often think they make monsters stronger by giving them more HP, more damage or more AC, but the only effect this has is that you get the same outcome, it just drags the fight out longer.
You give an Adult Dragon more actions, say one for each player and with each action they can attack 3 times. Now you have a real threat to a party of 4 characters of very high level, 5th level character stand absolutely no chance at all (as they shouldn't) and you don't need to change anything else about the dragon and the fight will be determined much faster.
Action Economy... its a thing.
The expectation is that the difficulty of a CR X for a level X party is consistent, and it's not. Two CR 1s vs a level 1 party is a legitimate scary fight. Two CR 20s vs a level 20 party is not (the fact that one CR X is supposed to be a meaningful encounter is a separate math problem that exists at every level).
D&D 5e (can't speak to 5.5e) has one of the highest GM workloads of any given TTRPG. It is literally the worst system to GM. Largely this is due to the high level of player focus in the design.
We've all had those moments where players choose to have their characters do something entirely off the wall. High level play makes that even more likely. To design fun and engaging content post level 12 (in my opinion) is a case of having to fix what the rules designers never did. I'm actually fresh from two 1-19 campaigns that have just finished, and moving onto new 1-20 campaigns. I also run one-shots and shorter runs of a whole load of different systems. Designing encounters that aren't trivial is no small feat with D&D.
I mean even the designers can't get it right. Vecna in Vecna Eve of Ruin, is an absolute pushover. Can be destroyed really easily and really quickly. He's meant to be a big, big bad and yet my players described it as 'anti-climactic doesn't cover it'. Even in that supposed Level 20 end encounter, Vecna is entirely on their own. No minions, no environmental effects, no lair actions, no cool tactical elements. Just 272 hit points of pure boring. If the people paid to look after the IP and design the game can't craft a cool high-level adventure, what chance do you think GMs have?
By contrast, the last Pathfinder game I ran to max level was insane and their Gamemastery Guide was so much better written than the DMG, and a larger array of good 'official' adventures that run through the full array of levels was available.
Heck, even looking at the new Tales of the Valiant Game Master's Guide should have WotC and their staff blushing in absolute shame. That GMG is so well written. Like I was chatting with other GMs as our coupon codes dropped and we literally tried to rip it apart, but just couldn't. I'm seriously looking forward to running a high level game of ToV (with no D&D content).
The problem of high level play is that most GMs who run to high levels like myself, run full campaigns level 1-20. If you didn't get on board with their campaign early you aren't going to get the chance to join in later. Those few GMs who can and do have the skill to run high level one-shots are then also so in demand that they almost don't even have to look for players. In short, the problem is the rules designers and policy people at WotC. They need to actually start considering the DMs who make their game actually possible. I was hopeful that the 2024 DMG might actually help with this problem but after seeing the mess they made of the 2024 PHB - that hope has gone away.
DM session planning template - My version of maps for 'Lost Mine of Phandelver' - Send your party to The Circus - Other DM Resources - Maps, Tokens, Quests - 'Better' Player Character Injury Tables?
Actor, Writer, Director & Teacher by day - GM/DM in my off hours.
As several people point out here, we don't see many high-level campaigns because they require two things to work: 1) a high level of DM skill to craft and run, and 2) players who employ their characters' incredible powers with restraint. It can be done, however. I just completed playing in a fellow DM's level 11-20 campaign. Fun and challenging.
How did we do it? The game took place in isolated locations, with urgency, that placed limits on resources and downtime activity. Many challenges were best solved through non-combat means, which honestly ran little differently than if we were first level: they required role-play, patience, and creativity. Granted, not every group of players will be on board for this sort of thing. And the DM had a lot of work to prep encounters, usually with large numbers of monsters and homebrew villains to test the mettle of our powerful characters. Believe it or not, we nearly had a TPK at level 20, even with fairly optimized characters and experienced players.
This experience made me want to run a high-level game. It's a thrill to employ 9th level spells or legendary capstone abilities, rather than just read about them.
I am working my way up on a campaign, up to level 13 now, and I think the problem is predictability of the party.
If you know the party already (IE they have worked up to higher levels) then it's going to be much easier to make balanced combats for them to deal with.
EG, if I use a combat calculator and make a series of deadly encounters for the adventuring day for 5 level 18 characters, on the assumption I will pick up players afterwards, so have no idea what characters I'm running for, then it will be impossible for me to know if they are a challenge. Let's say the first combat is a horde of smaller enemies with pack tactics. If the party has lots of magic users with area effects, then dang, that was easy. If they're heavy on rangers, rogues, and barbarians, then dang, this will be a challenge. Next encounter, they are all flying enemies. Welp, fighter and ranger have bows and endurance, barbarian is grumpy, and the spellcasters just did another big spell and made it easier. Now let's pit them against something with resistance to magic, and suddenly a spellcasting party is struggling... do you see the problem? You can't predict if the encounters will actually be difficult or easy, because at high levels there are ways to shut down entire swathes of tactics.
Make your Artificer work with any other class with 174 Multiclassing Feats for your Artificer Multiclass Character!
DM's Guild Releases on This Thread Or check them all out on DMs Guild!
DrivethruRPG Releases on This Thread - latest release: My Character is a Werewolf: balanced rules for Lycanthropy!
I have started discussing/reviewing 3rd party D&D content on Substack - stay tuned for semi-regular posts!
I fully agree, but I still hold that the action economy is the problem. The issue with a 1st level party facing 2 CR 1 monsters is that these have reasonably equal action economies. 20th level characters have a much higher action economy than CR20 monsters. A CR20 monster should be getting 2-4 Actions, potentially 7-10 attacks in a round. Than you have a fair fight. You don't need to change anything else than simply give them more actions in a round.
I have been doing it for years, it completely fixes the CR system. When a CR20 monster has 3 actions and can do 7-10 attacks in a round, 4 20th level characters are going to struggle to survive that fight. Its really that simple to fix.
This is true but only if you are trying to resolve balance exclusively by the rules. Meaning that if you are building encounters with the intent to run them as written, balance especially at higher levels is unlikely and even at lower levels its mostly coincidental.
Adjusting encounters to suit the party is just a part of the game, but its not an exclusive "thing" to D&D, this is a problem in most games.
The issue is that the advice for GMing high-level encounters is terrible. There is this consensus of upping AC, Hit Points and Damage and these things do not solve most problems because they are not the cause, they just drag out fights. A D&D battle should last 3 rounds, no more and those 3 rounds should be filled with nail-biting dice rolls. Give Vecna 3-4 actions per round. That's all you have to do to turn that encounter into a serious problem for any max-level adventuring party.
The reason this works is because downing a 20th level character in a D&D encounter is completetly meaningless. They will be back on their feat and taking actions the very next round and players can bounce of like this repeadly. A deadly encounter should be just that, a 50-50 in which players should have a clear expectation that some characters will die, their might be a TPK and the odds on their success are a crap shoot at best.
The meaning of deadly is that people will die everytime. If that doesn't happen, its not a deadly encounter.
This is, to my mind, a highlight of a separate issue - the pop-up combat mechanics. This is something I have worked to solve with my Gritty, Consequential and Impactful Dying Rules, which make it much more beneficial to heal people whilst they are alive than when they have already fallen! It also includes lingering injuries which, again, make people want to stay upright, rather than the current mentality of "don't heal me yet, wait until I die, it'll be more useful then!".
The problem as-is is that healing someone for 20hp when they are at 4hp, and them then taking 25hp and dying, is less efficient than letting them take 25hp whilst at 4hp, die, and then pop them up with 20hp of healing.
I do like the idea of the additional actions, but isn't this what Legendary actions do already? Plus lair actions for bosses on home-turf?
When I'm homebrewing I like to include the bosses main abilities as legendary actions. For example, I made the Doom Slayer for a oneshot, and one of his legendary actions was to charge the BFG. He couldn't fire it unless he charged it, so it was telegraphed in a legendary action. I also like to have at least one Reaction and one Bonus Action on each monster when I write them - too many have an action and that's it, with multiattack, making other actions worthless in combat.
Make your Artificer work with any other class with 174 Multiclassing Feats for your Artificer Multiclass Character!
DM's Guild Releases on This Thread Or check them all out on DMs Guild!
DrivethruRPG Releases on This Thread - latest release: My Character is a Werewolf: balanced rules for Lycanthropy!
I have started discussing/reviewing 3rd party D&D content on Substack - stay tuned for semi-regular posts!
Again that is not a problem, it is a feature. D&D 5e is not a challenging tactical combat game, it is a "hey remember that time we killed a dragon by dropping an anvil on it's head?"-game. It is designed for a group of friends getting together, eating snacks, drinking, and just chilling and having a good time. It is more akin to Skyrim than to Elden Ring. It deliberately doesn't require skill, smarts or tactical genius to play. It deliberately is almost impossible to die / TPK. It is not meant to be challenging, it is not meant to have a significant risk of failure, it is not meant to require teamwork and coordination. It is a power fantasy where players can make whatever character they want and have that character become a monster-killing superhero. It is meant to be able to be played by groups where nobody has a thorough understanding of the rules, and everyone's a little tipsy or on a sugar rush, or the whole party are 10 year olds.
CR is designed so that a party of randomly assembled 10 year olds with whacky characters designed to emulate their favourite superheroes can fight & kill a dragon after school. Or so a group of first-time gamers can get together after work at a pub, and play as their fantasy ideal versions of themselves. Or so that guy who says they want to play the "worst character" can still be an effective member of the party, get to occasionally be the hero and have a good time.
If the CR system was designed such that it lead to almost-TPKs even among experienced players with optimized characters with Legendary magic items, then it wouldn't work for casual players who just want to muck around and try crazy plans like throwing a vampire into the River Styx. See it is impossible to please everyone? So the game designers wisely choose to make the guide books designed for those casual players who don't know what they are doing so don't know how to bend / change the rules without breaking the game or who aren't going to be spending 4-5 hours prepping for a game, and left the serious tactical players to figure out for themselves how to make tactical challenging combats that suit their tasted. Because those first-time gamers or tipsy frat boys are going to need to refer to the books a lot more than the table of tactical geniuses who have spend hours studying the rules to make the most powerful character they can.
So, this is where I once again point to Wave Echo Cave as an example of the game as designed. That single dungeon is a great model of the game working and planned as intended. The lesser monsters throughout absolutely bleed away character resources. This is the reason that Neznar, the black spider, is so underwhelming. The desingers understood that the party were likely extremely low on resources by the time the reached that final boss. Assuming too, that the party only short rest twice in their run through those caves it can actually be a rewarding fight. It's never going to be an easy TPK for the black spider, but it will be a tough one for the party. That's the game working as designed. For years, I know I never ran that dungeon as intended with characters taking short rests whenever they felt like. It was only when limiting them to two short rests per in game day that Wave Echo Cave finally made sense to me.
The same frankly is true of high level play. There are underlying assumptions that the designers have made. Mostly they are carry overs from previous versions of the game. The long dungeon crawl is actually core to the assumptions made in the design of D&D 5e. The final villain's lair ought to be jam packed with enemies. It ought to be unsafe to take a long rest in. There is however, by any measure, a player entitlement issue with D&D specifically. And the designers have continually played into it. That's not a bad thing necessarily let's be clear. Giving players more options and more choices is generally a good thing. It is the implementation that becomes an issue.
Let's put together a hypothetical adventuring day for high level though as a thought experiment. We're assuming 5 adventurers here, all at level 19. Daily XP budget is 150,000 and we need to get 6 medium to hard encounters within that budget. This will be a dungeon that is undead themed.
Encounter 1: Boneclaw, Devourer 27,600/150,000xp
Encounter 2: Sword Wraith Commander, 5x Sword Wraith Warrior, 2x Revenant 55,100/150,000xp
Short Rest 1 happens here, as it's about 1/3rd of the way through the adventuring day.
Encounter 3: 5x Beholder Zombie, Deathlock Mastermind 80,900/150,000xp
Encounter 4: 2x Death Tyrant 115,400/150,000xp
Short Rest 2 happens here, as we assume we're now 2/3rds of the adventuring day done
Encounter 5: Nightwalker 140,400/150,000xp
This is where we can see the designers messing up the DMG. They didn't consider that at high levels a larger daily XP budget was going to be needed. We're effectively likely to go over our budget here with our final of the 6 encounters today. However, we can also see that the assumptions of the DMG are based on a party of 6, not of 5. (For me, 5 players is typical which is why I'm showing off a day for 5 players). So our XP numbers are slightly incorrect.
Encounter 6: Demilich 160,400/150,000xp
This is all medium difficulty encounters. No hard encounters here, all medium encounter difficulty. Now, I've actually run this adventuring day, and I did so with a Barb, Wizard, Rogue, Cleric, Paladin. No multiclass, no feats, and one magic item each. Even this adventuring day was fairly easy and fairly survivable for the party. In fact there is a massive gulf between this rather easy adventuring day, and an adventuring day that would lead to a TPK. This adventuring day ended with no-one having dropped unconscious, and only the Barbarian having lower than max hit points! See what I mean about easy?
Fact is that the rules designers just weren't great at designing the fundamentals of the game at the higher end. My assessment of such is that they have given too many shiny things to players, but not counterbalanced with cool monster effects. So, you're kinda left with only one choice as a GM - more, and difficult encounters designed to wick away resources to the point that they're almost tapped out by the time the reach the final boss. It's effectively a common limitation. By the by, I'd consider simply throwing more monsters along with the end boss to be part of this.
Simple answer to why so few people run high level though is because Crawford et. al suck at their jobs. The DMG is garbage.
DM session planning template - My version of maps for 'Lost Mine of Phandelver' - Send your party to The Circus - Other DM Resources - Maps, Tokens, Quests - 'Better' Player Character Injury Tables?
Actor, Writer, Director & Teacher by day - GM/DM in my off hours.
While it's absolutely possible to have a beer and pretzels D&D game, it's really not the way the game is designed -- if it was, we certainly wouldn't have concepts like the adventuring day (I assure you, tracking multiple different resources that can be depleted across multiple encounters and probably multiple sessions is not what casual gamers want to do) and high levels would be much simpler (casual players also don't want to shuffle through their 20+ prepared spells and 100+ available spells).
If you're not doing major reworks to the game system, that should mean that the encounter system should be designed to be lenient at the levels normally played by casual gamer (i.e. low level) and increasingly less forgiving as levels increase (as, presumably, people who play at higher levels are comfortable with higher difficulty).
Citation needed. When I looked through the DMG all I could find are:
"Party Size: The preceding guidelines assume that you have a party consisting of three to five adventurers" - So your party is at the upper end of power due to the large number of players,
Note that:
"Increase the difficulty of the encounter by one step (from easy to medium, for example) if the characters have a drawback that their enemies don’t. Reduce the difficulty by one step if the characters have a benefit that their enemies don’t." - This includes surprise, so if your party managed to get surprise your encounters were not "Medium" difficulty they were Easy.
Also note that:
"This adjusted value is not what the monsters are worth in terms of XP; the adjusted value’s only purpose is to help you accurately assess the encounter’s difficulty."
and:
"Use the Adventuring Day XP table to estimate how much XP that character is expected to earn in a day."
Your example Adventuring day only earns the party:110,300 XP / 150,000 XP Budget. -> i.e. an Easy-ish adventuring day which is exactly what you observed.
Two sentences later we see "This provides a rough estimate of the adjusted XP value for encounters the party can handle before the characters will need to take a long rest." So no, he's doing it correctly.
OK! But this time let's do it properly.
Total Daily XP budget for monsters = 150,000 XP
Using the same monsters as you this was my adjustment to get the appropriate total:
Boneclaw = 8400 XP * 2 = 25200
Devourer = 10000 XP = 10000
Sword Wraith Commander = 3900 XP = 3900
Sword Wraith Warrior = 700 XP *5 = 3500
Revenant = 1800 XP * 2 = 3600
Beholder Zombie = 1800 XP * 5 = 9000
Deathlock Mastermind = 3900 XP = 3900
Death Tyrant = 11500 *3 = 34500
Nightwalker = 25000 = 35000
Demilich = 20000 * 2 = 40000
Total: 150,200 XP
Now to assemble them into encounters:
Encounter 1: Sword Wraith Commander, 5x Sword Wraith Warrior, 2x Revenant = 22,000 = Medium encounter
Encounter 2: 2x Bone Claw + Devourer = 39,300 XP = Hard encounter
Short Rest 1 happens here, as it's about 1/3rd of the way through the adventuring day.
Encounter 3: 5x Beholder Zombie, Deathlock Mastermind = 25,800 = Medium encounter
Encounter 4: 3x Death Tyrant = 69,000 = Deadly encounter
Short Rest 2 happens here, as we assume we're now 2/3rds of the adventuring day done
Encounter 5: Nightwalker = 25,000 = Medium encounter
Encounter 6: 2x Demilich = 60,000 XP = Deadly encounter
I'd be super curious what the results would be if you reran the test with this appropriate adventuring day based on the DMG rules. But again make sure they don't have any magic items as magic weapons negate one of the Demilich's (and some of the other monster's) resistances which are a core part of its defenses. - Though honestly the Demilich is pretty terrible, you need to have that encounter in a tiny room or hallway in a tomb, I'd replace it with a Lich + 3* Bodaks if playing in a large arena.
It won't change, because he used the rules correctly. Also, using your interpretation, bear in mind that by your logic this is an accurate 1st level adventuring day for 5 PCs (daily budget: 300x5 = 1500)
What do you think would happen to a level 1 party that actually tried to get through that?
Just a couple of comments since I have nothing to do with the OPs game. However, in this case it is a level 25 encounter. What makes you think that any of these feeble ideas would have any reasonable chance of success?
1) Wind walk + pass without trace. Wind Walk doesn't make you invisible. Those oddly moving clouds in the sky are likely to stand out no matter what else is happening especially if the castle is alert. Pass without Trace does NOTHING if you are visible. So ... they will get noticed and wind walk doesn't make them invulnerable so they get to be targeted by whatever defenses can aim towards the sky. In addition, the players were unaware the castle is actually a sentient creature that is likely to notice ANYTHING entering.
2) Mounts reduce your time on the field and get you closer faster but don't really do anything else. They also die quickly. Also by that level there are likely characters with base movement and magic that are just as fast. Running in quickly is likely a non-issue and good luck keeping out an eye for traps dashing towards the castle at 160'/rd. I'm assuming the kind of traps found at L25 are likely nasty enough that avoiding them would be preferable to running over them.
3) Inconspicuous mice that can be seen by detect magic. Which mundane pets can see and attack ... trained raptors in the air. Keep in mind that the sewer pipe you are talking about is a part of the sentient castle. Interesting when the mice are swallowed and confined.
4) True Seeing is common at that level. See Invisible itself is only a level 2 spell. Against opponents of this nature one would expect most of the guards to be able to see invisible and I would expect the castle itself will have both true seeing and blindsight.
5) Seeming is an illusion and is bypassed by True Seeing which at least some of the guards would be expected to have. Minimum.
It seems possible that you've never played or run a game for high level character and the shenanigans that they can get up to. My home game is currently level 18 and the Adventurer's League run of Out of the Abyss I am running is at 12th. I've also played a fair bit at tier 3 and a few tier 4s - so I've seen what can be done and also what a DM will have in mind that counter's all the suggestions you made above unless the party does something more than that.
6) This assumes that the enemy gives you a target and there isn't some reason you actually need to get close ... like rescue a princess?
The DM doesn't need to ban anything. The players need to come up with different plans since the low level tactics require modifications to be successful. Mind blank, the spell to mask magical signatures, perhaps finding a herb to mask your scents - LOTS of things to do but the simple cast a spell and solve the problem is usually off the table.
They "thought they had won" ... this doesn't necessarily mean they marched through and cleaned out the castle ... it may have meant they silenced the wall defenses and thought they just needed to enter the castle and mop up. Ooops ... DM had additional twists in mind. My guess is that the sentient castle is just the next line of defence after the walls and there are likely more. I I was running it, I'd guess that the princess is on a Demiplane and that if they get through the castle they can find a portal or some method of access to the next part of the adventure ...
I'd suggest that there may have been other methods possible but none of the ones you suggested were at all likely to work in a level 20+ game.
On the matter of Short Rests: Creating Adventures - Dungeon Master’s Guide (2014) - Dungeons & Dragons - Sources - D&D Beyond (dndbeyond.com)
'In general, over the course of a full adventuring day, the party will likely need to take two short rests, about one-third and two-thirds of the way through the day.'
On the matter of adventuring day: Creating Adventures - Dungeon Master’s Guide (2014) - Dungeons & Dragons - Sources - D&D Beyond (dndbeyond.com)
'Assuming typical adventuring conditions and average luck, most adventuring parties can handle about six to eight medium or hard encounters in a day. If the adventure has more easy encounters, the adventurers can get through more. If it has more deadly encounters, they can handle fewer.'
On the subject of your 'proper' encounter
You have added in Deadly encounters which automatically skew the number of encounters we should have. Pages 81-84 of the DMG (first Printing, December 2014) are where to look for this information. If you're throwing in 'deadly' encounters it means you need fewer encounters during the adventuring day. If you follow the assumptions laid out. Now, without actually running the encounter, I'm fairly sure that even what you've proposed is pretty easily survivable.
At best guess, even assuming the enemies always get surprise, the party survive easily. Now, I will actually run this adventuring day with my group of GMs (we playtest stuff quite regularly these days), but I guarantee that none of these encounters will run for more than 6 rounds. For the Demilich, consider that the barb might well have functions that turn their attacks into magical attacks. Paladin's smite will cleave through some of those immunities and resistances too. We have both Cleric and Paladin represented here which gives some benefits against undeaed too. In this instance I've picked a quite difficult, if basic team of adventurers for enemies to go up against. I would even go so far as to claim that even with a team of just four - this adventure day is totally survivable. Probably pretty easy too. Based off experience, even losing the Paladin from the equations, the party aren't going to have a single member fall unconscious.
I did use the rules correctly, and would suggest taking a look again at the Encounter section of the DMG.
DM session planning template - My version of maps for 'Lost Mine of Phandelver' - Send your party to The Circus - Other DM Resources - Maps, Tokens, Quests - 'Better' Player Character Injury Tables?
Actor, Writer, Director & Teacher by day - GM/DM in my off hours.
However you work it play at tiers 3+ can be difficult for a few reasons - 1) characters are extremely capable and players are generally experienced to highly experienced so they are capable of lots of out of the box solutions you may not be prepared for - no matter how prepared you are. 2) the CR system tends to be out of whack at high level so it takes more DM work to make it work - and DMs don’t all want to do or have time to do that much work on top of any/everything else they are already doing. 3) making the sorts of adjustments as well as prepping for the shenanigans of the players takes experience to do well so everyone enjoys it. Most DMs have under 10 years experience (many less than 5) mostly at lower levels of play so they tend to be unprepared for this. Only 3.x and 4e made any efforts at higher level play. 3e had the epic players guide (essential for post L20 play and a decent starting point for high level play. 4e allowed PCs to progress to L30 which covered the vast majority of tier 3+ play. Even in those high level play was problematic for many of the reasons given above. To do tier 3+ campaigns basically calls for experienced players and highly experienced DMs and while there are lots numerically given the number of folks playing there aren’t that many in most locations so you don’t see that many games. Even for those that are most of our games are at lower levels as those can be just as fun so why go to the effort?
Wisea$$ DM and Player since 1979.
David42 is absolutely correct a few points for clarification then I will let this go as the notion that this stuff did not occur to a group of seasoned gamers is incorrect.
A very high percentage of random mobs and every mini boss or boss in a l25 game should be assumed to have most if not all the following - Alert, True sight, Multi attack, Cast + Melee same round with various attacks, no disadvantage for casting within 5', Fly on demand, Tier 1 and Tier 2 Epic Spells, various resistances. Passive Insight, Perception and Investigation 25+, to hit and save modifiers of 12+, outright immunity to 1 damage type. Potentially reflective damage, potentially healed by one type of damage.
There could have been traps, a horde of invisible mobs lying in ambush between us and the castle, or anything that the dm could have dreamed up - proceeding cautiously is always best. We cleared the field - the sorcerer animated some boulders and had them go in front of us as cover- we took some damage on the way in but where at least reasonably certain there was nothing behind us then proceeded to deal with the guards, ballista's, mini bosses etc.
In this case these mini bosses had broad resistances - some times its elemental only - sometimes its force - sometimes its whatever random combination of resistances ----as a group the key is to figure out what they are resistant to then leverage that. See what saves they are bad at - then leverage that etc etc its not a low level game with big numbers you literally have the entire PHB DMG and Epic ruleset at your disposal the level of creativity possible is staggering.
Plane hopping - world hopping - continent hopping all happen on a regular basis my guess is that when we kill the castle the force that animated it will be dispelled and the princess will be inside - then again maybe not the DM made it up so its his story to tell. Likely the evil force that animated the castle is a minion of the ultimate threat and the next task will be to go hunt that down.
As an aside much has been made of the CR system and action efficiency - in most high level games it is assumed that the fight difficulty is adjusted by the DM prior to each session given the groups overall strength--- a combination of AC - spike damage - continuous damage - legendary skills - lair skills - high dc saves - and unique homebrew skills and spells all blend together to make a fascinating adventure.
This is a very basic example but illustrative of my point --- you kill a group of 4 seemingly normal fire elementals as soon as the 4th dies the embers from all 4 reform in to 2 xl fire elementals kill those 2 then the embers reform into an xxl elemental - its a common tactic and makes parties be cautious about not dropping everything on 1 part of a fight. Assuming that it is all correctly balanced it makes for a thrilling fight that the party has a reasonable chance of winning.
There has been a lot of great discussion in this thread I will just add that I hope more folks get motivated to play in and run these games.
I think Flee Mortals is better, but I have still had issues with those monsters not being a threat.
Pretty sure Flee Mortals tries to obey 5e CR rules, just make them more interesting, so it won't solve any simple difficulty problems.