D&D 5E Campaign Pacing

[Li Shenron]

I wouldn't overdo this. AD&D2 in particular had a bunch of levels in the middle where it took ages to level up, so much that player characters ended up feeling more or less static. I would not mind this in any other RPG, but D&D comes with the expectation that characters level up every so often.

If it worked differently in AD&D then maybe the expectation is rather on current RPG editions than the whole history of D&D?

Anyway, I have very little experience of AD&D, but at least I know that in older editions there were also many "empty levels" in classes, where the character would advance only with more HP and perhaps better attacks, but no new tactical abilities. The new default in 5e is to always grant something new at level up (although sometimes the 'new' ability granted is still only a numerical increase somewhere).

Unfortunately, I haven't had the chance yet to play 5e until high levels... I have the feeling that the situation is now much better compared to 3e, mostly because the number of spells known by a character is much smaller. But I can say that in 3e the standard XP advancement was definitely a problem in my groups, as we always ended up not even having time to try out the new abilities, and already getting more on top of them.

 

[Krachek]

Heh. That was for illustrative purposes and it has prompted the intuition I hoped. We likely share a feeling that a 1:1000 chance of dying is too low / feels too easy.

But the maths proves that over the number of encounters that characters will face, even a 1:100 chance of dying is going to kill most of them. I believe that points toward the role of revival magic - Gentle Repose, Revivify, Raise Dead, Reincarnate, Resurrection, True Resurrection. In another thread, I asked DMs their experience with death revival and many related very few deaths, and very few or no revivals. I take no issue with that - of course - but in my opinion the mechanics of D&D point in an entirely different direction: a greater chance of death, and a greater chance of revival.

I would argue that

a) In order that encounters feel challenging and tense, the risk of death (or other great loss) must be palpable, and
b) The lethality ratio that equates with "palpable" will be great enough that characters are certain to die over the expected number of encounters
c) Therefore, something needs to bring them back again - a revival ratio

Formulae

1-(1-lethality)^encounters = death chance

1-(1-lethality*(1-revival chance))^encounters = permanent death chance

I may post up a table later showing the implications (for varying lethality and revival assumptions) in terms of how many characters need to be created in order to have one survive across each tier boundary. It is possible to pick tuning values for lethality and revival that match our world assumptions for density of character class equivalent individuals at each tier.

I found that the following lethality and revival assumptions worked quite well

Deadly encounter = 1:10 chance of death (of 10 characters going into a deadly encounter, 1 dies)
Hard encounter = 1:100 chance of death
Easy-Medium = attritional (no chance of death when considered separately)

Revival per tier
Tier 1 = 1:5 (of 5 characters that die, 4 are permanent)
Tier 2 = 2:5
Tier 3 = 3:5
Tier 4 = 4:5

The scaling revival likelihood does two things:

1) maps quite well to a world assumption that however many individuals are tier 1, there are an order of magnitude fewer per tier upward. About half being assumed to retire, and all but one of the rest die.
2) appropriately acknowledges the greater work and narrative significance invested into a higher tier character

I do understand your point.

But human mind is a strange thing.
The goal is to perceive a meaningfull challenge, despite of the 1:1000.
The more you focus on the mechanics, and your mathematical chance of success, the more your mind will reduce the stress and the win will become a boring certitude.
The illusion of challenge, or the illusion of the whole DnD world is rather fragile. If you search for flaw and contradiction you will rapidly discover the truth. It is a game, with a fool setup.
But if we want and help to keep the illusion it will be quite effective.

 

[clearstream]

I wouldn't overdo this. AD&D2 in particular had a bunch of levels in the middle where it took ages to level up, so much that player characters ended up feeling more or less static. I would not mind this in any other RPG, but D&D comes with the expectation that characters level up every so often.

On the other hand, I agree 5e levels up way too fast. It's so temptingly simple to just divide xp awards by 10.

When I worked through the numbers, doubling the RAW assumption looked quite good. I defaulted to that in my OOTA campaign, and am thinking of sticking with it.

2 sessions 1-2
4 sessions for 2-3, 3-4, 4-5
6 sessions for 5-6, 6-7, 7-8, 8-9, 9-10, 10-11
6 sessions for 11-12, 12-13, 13-14, 14-15, 15-16, 16-17
4 sessions for 17-18, 18-19, 19-20

102 four-hour sessions to level from 1-20, of which about 70 are spent in the 4-15 sweet spot. Two years if you play once a week, or a year for keener groups. Twenty-five days for the very keenest (allowing 8 hours each day to sleep, eat and occasionally bathe, which could be thought too much).

 

[clearstream]

I do understand your point.

But human mind is a strange thing.
The goal is to perceive a meaningfull challenge, despite of the 1:1000.
The more you focus on the mechanics, and your mathematical chance of success, the more your mind will reduce the stress and the win will become a boring certitude.
The illusion of challenge, or the illusion of the whole DnD world is rather fragile. If you search for flaw and contradiction you will rapidly discover the truth. It is a game, with a fool setup.
But if we want and help to keep the illusion it will be quite effective.

Agreed. If we learned nothing else from the Wizard of Oz, we learned that it's important not to confront suspension of disbelief with the wires and pulleys behind the screen. (Also, if you are a wicked witch, don't stand under any buckets. That's probably less relevant here.)

That's why this is all secret-squirrel, for DMs only...

 

[BookBarbarian]

Very good analysis!



This confirms my personal idea that that the game pacing in 5e is still too fast for my tastes.

I generally prefer* a somewhat logarithmic progression of level rather than linear, meaning that I'd like it to take always a bit longer time to get to the next level, compared to the time it took to reach the current level from the previous. This is something that is actually shown by the XP thresholds to level up (which increase faster than linearly), but is then neutered by the XP awards by encounter level (which also increase), together ending up with a mostly linear level advancement by time.

*if you wonder why I prefer that, there are multiple reasons:
- is feels more realistic, in the same way as in real life it takes a short time to learn the basics of anything (e.g. learning some chords on a guitar enough for a few songs) and then progressively more and more time for further improvements
- it gives the feeling that playing at higher level is actually more demanding than lower level, and thus higher level suits better players
- it gives more time to the players to adapt to the higher complexity of higher levels, instead of throwing them onto the next when they still haven't grasped the abilities acquired at the previous

Interesting. If higher levels progressed slower I would be very worried about campaign fatigue. Even if they progress quickly I'm still worried about it.

While I think 50ish sessions to level 20 is a great rule of thumb, 30ish adventuring days is ridiculous to me. However with liberal use of downtime, or Adventure's in Middle Earth's Fellowship Phases, I can make it 30 adventures rather than adventuring days. 2-3 adventures a year would mean 10-15 years to get to level 20. Which seems ok to me.

So I guess I'm more worried about the narrative aspect of it, than the mechanical.

 

[iserith]

Some data from my recent campaigns:

The Delve, a town-to-dungeon campaign, saw characters go from 1st to 8th level in 20 sessions. Because of a greater number of players than seats per session and at least two PCs per player, the cast rotated around quite a bit which would seem to dilute the XP pool. However, it looks to be more or less on track with your projections. I don't have any hard numbers I can share on number of encounters or difficulty.

Host of the High Chieftain, a rewrite of Red Hand of Doom, ran 16 sessions and the characters went from 6th to 9th level. So that one seems to be off from your projections. Here it was also a player pool, albeit a smaller one than The Delve plus in the latter quarter of the campaign all comers were welcome, and everyone had 2 PCs (only one active at a time). Standard XP rewards, but your backup characters got half the XP your active character got. I also can't share any hard numbers on number of encounters or difficulty. I wonder if the extended number of sessions had something to do with a fair amount of mooks being in encounters which didn't particular increase XP or difficulty much but did increase the time to resolve a scene. It's also possible that the characters were close to 10th level, but I didn't give XP for the last session because we finished the campaign. That may bring it closer in line.

Summoning @Valmarius, @Lanliss, and @mexicangringo for comment if they are so inclined.

In any case, it's an interesting analysis and thanks for posting it. I've shared it with some other DMs.

 

[Tony Vargas]

One of the interestingly entangled aspects of 5th edition are the guidelines on Campaign Pacing. It's elements are short rests and long rests (PHB 186), adventuring days (DMG 84), encounter XP thresholds (DMG 82), creature CR (MM 9), sessions (DMG 261), XP and levels (PHB 15). Character resources are tied to it.

Downtime days?

So I became curious about the implications of how it all joined up, for example how much XP worth of encounters are expected per session, and what might that mean for rests?

it takes 33 adventuring days to level from 1 to 20
...

Yep.

You also spend more days/encounters/sessions at each level from 4-10 than 1-3 or 11-19 (I suppose you could spend as much time at 20th as you liked.)
That should maximize time in the most-playable 'sweet spot' levels.

In contrast, as I remember AD&D generally playing out, you'd have a long hard slog to get to 2nd, speed through 3-7 (the sweet spot), and hit a wall sometime after 9th depending on class. Just the opposite of what'd maximize fun. :shrug: 5e learned something, I guess.

Once we expect 144 encounters, we can make a statement like if a PC has a 1:100 chance of dying per encounter, then they're ~76% likely to die before 20th level. A 1:1000 chance produces ~13% chance to die before 20th level, which sounds fine until you recall that a party has four characters. In fact, I believe this gives us a big hint that allowing revival spells to play a role is a pretty good idea. But that's really a derail (into ToA!)

Interesting. "Too easy," indeed....

 

[Li Shenron]

Interesting. If higher levels progressed slower I would be very worried about campaign fatigue. Even if they progress quickly I'm still worried about it.

While I think 50ish sessions to level 20 is a great rule of thumb, 30ish adventuring days is ridiculous to me. However with liberal use of downtime, or Adventure's in Middle Earth's Fellowship Phases, I can make it 30 adventures rather than adventuring days. 2-3 adventures a year would mean 10-15 years to get to level 20. Which seems ok to me.

So I guess I'm more worried about the narrative aspect of it, than the mechanical.

Yes, it's definitely more about the narrative! I have NPCs that have been around for centuries and I won't have every band of PCs catch up with them in a month.

 

[clearstream]

Some data from my recent campaigns:

The Delve, a town-to-dungeon campaign, saw characters go from 1st to 8th level in 20 sessions. Because of a greater number of players than seats per session and at least two PCs per player, the cast rotated around quite a bit which would seem to dilute the XP pool. However, it looks to be more or less on track with your projections. I don't have any hard numbers I can share on number of encounters or difficulty.

Host of the High Chieftain, a rewrite of Red Hand of Doom, ran 16 sessions and the characters went from 6th to 9th level. So that one seems to be off from your projections. Here it was also a player pool, albeit a smaller one than The Delve plus in the latter quarter of the campaign all comers were welcome, and everyone had 2 PCs (only one active at a time). Standard XP rewards, but your backup characters got half the XP your active character got. I also can't share any hard numbers on number of encounters or difficulty. I wonder if the extended number of sessions had something to do with a fair amount of mooks being in encounters which didn't particular increase XP or difficulty much but did increase the time to resolve a scene. It's also possible that the characters were close to 10th level, but I didn't give XP for the last session because we finished the campaign. That may bring it closer in line.

Summoning @Valmarius, @Lanliss, and @mexicangringo for comment if they are so inclined.

In any case, it's an interesting analysis and thanks for posting it. I've shared it with some other DMs.

I'm 21 sessions into OOTA. There are six PCs in the party. Four of those are closing on 6th level. We're about to hit the mid-point of the campaign narrative arc.

Thus my pacing has being much slower. Partly that is down to a large party: it takes much longer to play through each encounter, but XP to each character from the encounter divides more ways and stays the same. I plan to pick up the pacing to hit about double the guideline sessions per level.


[Edit: 21 sessions. So hitting the pacing I want (double guideline) pretty well. Don't know where 31 came from?]

 

[Stalker0]

So if I understand this correctly, in a standard adventure a player will level before they even take a single long rest (average a little under a day to level). Now...that is quick!