Uncertainty in RPG Worlds

Uncertainty - where even the GM does not know, has not defined, the truth of the matter.
I find this dramatically very powerful.
For instance, in my Wilderlands, two young lovers, Cassandra the Druidess & Polyachus the Apollo priest, both fairly major NPCs. and Cassandra a very long term NPC from a previous campaign, had taken shelter on a remote island to escape the evil Archmage Oriax. Unfortunately that island was later occupied by evil Skandik pirates, and it was assumed the lovers had met a terrible fate. But when the Skandiks had left, and others arrived on the island, they found only beautiful songbirds, never seen before. The legend grew up that the gods had taken pity on the lovers and transformed them into birds, and the birds now seen on the isle are their offspring.
I have no intention of ever proclaiming - or deciding - a definitive truth of the matter.
I do this with other Big Mysteries of the campaign, such as whether or not the gods really exist. I have my suspicions, but have no intention of forming a definitive answer.
I find this approach makes the campaign world feel much more real to me. I've noticed that some other designers use this sort of ambiguity, eg in the Elder Scrolls CRPGs, as a way to create depth and mystery.
Anyone else do this? Is it a common practice?

5e - Slowing advancement after Level 10

The 5e XP table is designed to speed up advancement after level 10. Discussion on doing the opposite here for a megadungeon sandbox campaign with PCs of various levels. Conclusion is that simply halving XP awards to level 11+ PCs should work best, giving the following effective requirements:

10 to 11: 21000 > 21,000 XP as per RAW
11  to 12:15000 > 30,000
12 to 13   20000 > 40,000
13  to 14  20000 > 40,000
14 to 15   25000 > 50,000
15 to 16   30000 > 60,000
16 to 17   30000 > 60,000
17 to 18   40000 > 80,000
18 to 19   40000 > 80,000
19 to 20   50000 > 100,000

Going Epic-10 in 5e D&D

I suggest to DMs that if they are running a story-based, low combat game in a lowish-fantasy setting, then go “E10” – levels should be capped at 10 not 20, and use the Epic rules in the DMG from 10th level: every 30,000 XP above level 10, a PC either gains +2 attribute (I’d retain cap at 20, maybe 22); or if the game uses Feats the PC can gain a Feat of the player's choice.

Character Advancement
Experience Points Level Proficiency Bonus
0                               1                +2
300                           2                +2
900                           3                +2
2,700                        4                +2
6,500                        5                +3
14,000                      6                +3
23,000                      7                +3
34,000                      8                +3
48,000                      9                +4
64,000                    10                +4
94,000                    E1               +4
124,000                  E2               +4
154,000                  E3               +4
+30,000                  +1               +4

Edit: The Primeval Thule 5e setting seems particularly suitable for this approach.

A Fighter Attacks house rule for OD&D to 1e/2e AD&D

I don't much like the Fighter ability to get his level in attacks/round vs 1 hd (or 0-level in 1e) foes, since it requires the GM to tell the player how many hd the enemy have, or the player has to constantly ask. It's also a bit silly at higher level IMO. But without some kind of multi attack Fighters can be seriously underpowered at higher level. I'm also not a fan of weapon specialisation bonuses to hit & damage, they change the balance of weapons and make two-handed high damage weapons underpowered; it also favours missile weapons with a high attack rate.

Somewhat influenced by 5e's insight that you can give staggered attacks/round bonuses to fighters without breaking the game (in 5e at 5th, 11th & 20th) matching where other classes like wizard/MU get big spell boosts, and wanting to restore the importance of Hero (4th) & Supehero (8th) levels, I came up with the following, specifically for my Swords & Wizardry game but it should work in all pre-3e D&D, especially if 18%-strength is not in play - so ideal for OD&D and BX-BECMI.

Fighters get +1 attack to the weapon normal attack rate at the following levels: 2nd, 4th & 8th. For melee weapons this gives an attack rate as follows:

Fighter
Level #Attacks
1 1
2-3 2
4-7 3
8+ 4

Fighter subclasses such as Paladin & Ranger (& Cavalier, Barbarian etc) get bonus attacks at 4th & 8th only, giving the following attack rates for melee weapons

Paladin & Ranger
Level #Attacks
1-3 1
4-7 2
8+ 3

So far this seems to work really well. I've tweaked Ranger bonus damage vs giant class creatures to take account of the increased number of attacks:

Ranger Level 1-5 Bonus damage +1/level
Ranger Level 6-15 Bonus damage +1/2 levels:
Level Bonus
7 +6
9 +7
11 +8
13 +9
15 +10 (max)

MUs and Thieves get +1 attack at 8th.
Clerics get +1 attack at 7th.

Update: polite, intelligent conversation here. :D

Running O5R - OSR in 5e

"Hello, my friend! If you seek knowledge about mighty Talos er running OSR in 5e, you have most certainly come to the right person."

Things to be aware of:

5e is a loose enough system that you don't really need to worry about balance any more than you would when running an OSR system. Players should know that running away is an option, and the GM should normally make this possible, possibly with some Athletics and/or Stealth checks to evade pursuit.

1. Generally you should just use the 5e monster stats. Some 5e monsters are a bit tougher than in OSR, eg orcs, some are weaker, eg manticores. But overall it works fine.
2. For monsters that don't exist in 5e, you can either use the closest 5e equivalent (tweaking a bit, especially any special powers) or convert the stats. For converting BFRPG to 5e I recommend:

AC: deduct 1 from the BFRPG AC since it starts at 11.
Hit Points: Usually doubling the BFRPG/OD&D hp works
Damage: Usually should be about double BFRPG number. You can often get this by keeping the BFRPG damage dice and adding an attribute bonus
CR: I pretty much eyeball CR, but halving the hit dice probably gives a reasonable number.
Attributes: One pretty much has to eyeball these with reference to 5e norms, eg low level monsters typically have a +1 to +3 attribute bonus for their attack attribute, mid level +4 to +6 (NPCs usually cap at +5/attribute 20), high level +7 to +9, rarely +10 (needs a 30 Attribute).
Attack Score & Magic Save DC can be calculated off CR & Attribute bonus, the totals should typically be around +4/DC 12 at low level, +7/DC 15 at medium level, +10/DC 18 at high level.

Monster numbers: 5e monsters can be more dangerous in large groups, though I've not found it a big enough issue to worry about generally. Most BFRPG adventures don't need numbers adjusting IMO, but Chris Gonnerman (Morgansfort) has a tendency to put 28 kobolds in a room, in a 1st level adventure. You may want to simply halve his monster numbers where they look excessive - but I'm running it in Swords & Wizardry and doing that, it's not a 5e thing so much as a designer thing.

I always use the Basic D&D 2d6 Reaction & Morale rules, I'd recommend using those or similar.

Treasure: I don't normally alter the BX/BFRPG listed treasure values unless I also reduced monster numbers (eg halve monsters, halve treasure). The main thing is don't assume the PCs will find the treasure, it should be an uncertain reward. It won't break the game if they are successful and find lots of treasure. Just don't let them buy unlimited magic items from the DMG, 5e isn't designed to support that. A limited list of Common & Uncommon items at twice creation cost is fine, eg I have a list of Uncommon items the PCs can purchase at 1000gp each. Likewise if you allow magic item creation at list price, enforce crafting time, level limits, formulae etc so they are encouraged to focus on the cheaper, easier stuff.

Aside from magic items, it's great IMO if the PCs find lots of treasure and are motivated to do interesting things with it. It definitely won't break the game (or any edition of D&D except 3e/PF, really).

Starting an RPG Meetup

From https://www.therpgsite.com/showthread.php?39280-Starting-a-D-amp-D-Meetup

CRKrueger asked:
"Hey, how about starting a thread detailing everything you did? Kind of a "Here's how you set up a successful Meetup" kinda thing."

Well, I didn't start cold. I had joined the local D&D Meetup in 2008 and become an assistant organiser there ca 2010, keeping that role with just one break of a year or so, until I started my own Meetup in early 2018. For a long time I was the one doing the scheduling and I had an eye on what worked and was worth emulating.

I was running a group at the big Meetup and we had a spinoff group playing beside us in the same Wilderlands campaign setting. The initial decision to fork was mostly due to lack of space at the venue (plus some issues with how the big Meetup asst org was running that day); we moved our two groups to another nearby pub venue I knew - this required some negotiation with pub management to get us a reserved space (which we may be losing, the new manager doesn't seem so keen) and added a third Wilderlands group. By this time we were basically a mini Meetup within the Meetup. When the asst org wanted to assert more control over us I decided to take it independent and start a new Meetup. I had already established a Facebook group for us which made communications easier - I was coming in with around 20-22 players.

For the new Meetup:
1. I secured a pub venue with good transport links in an area gamers could easily get to, that I knew was safe and reasonably pleasant.
2. I wrote a description for the new Meetup that made it clear we were open to new players.
3. I designed the Open Wilderlands multi-table campaign to be the core of the Meetup and to be very accommodating to new players coming in and out, no pressure. And no disruption/blame. I was influenced by Ben Robbins, Justin Alexander, and what I knew of how Gygax and Arneson had run their Greyhawk & Blackmoor campaigns.
4. I used the Meetup advertiser tools which announce new Meetups. Maybe because we immediately had over 20 members on the first day, their algorithms spread it further. Currently after 6 months we have 131 members, something which took the original D&D Meetup a few years (it's now the world's largest with over 4000).

I guess my advice would be:
1. Come in with some other people at the start. Money makes money and players make players. The main obstacle to others joining your Meetup is the apprehension that you're a lone weirdo with no social skills. I'd say you want at least an established play group (say 5 people) and at least one other person willing to GM as the Meetup grows. You should be normal, pleasant people - you will get the odd weirdo joining later, that can't be helped, but you need a good quality starting cadre.
2. Secure a safe, pleasant public venue with good transport links, at a good time of day when things are otherwise quiet (I chose Sunday lunchtime). Not all venues require booking, and indeed I think at the pub our first games did not have the space reserved, but I convinced the pub we were worth having.
3. It should be oriented around allowing drop-in & irregular play by players (I've resisted one GM's efforts to make it more restrictive), combined with very reliable GMing, which for me means low-effort GMing. The WotC hardbacks are not great for this, Phandelver is ok. Megadungeons are ideal; short wilderness delves in a sandbox are good too. But the main thing is that the GMs are reliable, flaky GMs will destroy a nascent Meetup.
4. Likewise the initial game systems used should be popular and accessible - making it explicitly a "5e D&D Meetup" helped a lot I think; you can always allow other systems later. 5e D&D attracts lots of new (often younger) players very keen to play, and without many preconceived notions or demands. And newbies attract newbies in a virtuous circle.
5. Know what you're doing - the more experience you have of how a public Meetup works before starting your own, the better. Eg I use the RSVP limits to set how many players can come each week, fitting the number of players to the available GMs. I allow people to bring 2 Guests, but not to bring in an entire established group just looking for a venue. The groups need to stay open to new players.

5e D&D XP Awards

After playing 5e D&D awhile, I settled on a nice simple metric for XP awards that works for quests & missions, social encounters, and exploration: my guideline XP award per PC is 100 XP per level. That's threat or accomplishment level, which may be the same as PC level but may also be higher or lower. For instance, the awards on the second level of a dungeon will typically be 200 XP/PC, while on the 9th level they'll be 900 per PC. An act worthy of 1st level PCs such as a minor village quest gets 100 XP/PC, while an act worthy of 10th level characters such as victory in a baronial war gets 1000 XP/PC. Saving the world type stuff worthy of 20th level characters gets 2000 XP/PC.

I normally cap any purely social activity awards at 1000 XP per PC per hour of tabletop play, or per 2-3 hours of Internet text-chat play, no matter the PC level, but really major social achievements such as forging an alliance or becoming monarch may get their own awards in addition, again at 100 XP/level per award.

For a 2-3 hour Internet session where things go a bit slowly, a useful guide is to award the standard amount of at least 100 XP per level of each PC, assuming they are actually doing stuff, up to 1000 XP/PC at level 10+.

For a tabletop session I'd recommend three times that a guide to a typical session's total award, ie:

PC Level / Individual Award / Typical Session Award / XP & Sessions to Level

1.        100 / 300  / 300 - 1
2.        200 / 600 / 600 - 1
3.       300  / 900  / 1800 - 2
4.       400 / 1200 / 3800 - 3 or 4 (total 7-8 to 5th)
5.       500 / 1500 / 7500 - 5 
6.       600 / 1800 / 9000 - 5 (total 17-18 to 7th)
7.       700 / 2100 / 11000 - 5 or 6
8.       800 / 2400 / 14000 - 5 or 6
9.      900 / 2700 / 16000 - 5 or 6 (total 32-36 to 9th)
10.     1000 / 3000 / 21000 - 7

Using this metric, a PC can expect to climb from 1st to 11th in around 39-43 sessions of play, somewhat faster than Gygax's '1st to 9th in 50 sessions', but not wildly different, and the main change being the fast levelling up to 4th or 5th level. 

It's ok for awards at low level (1-4) to be somewhat lower, about half the above, but 5e seems to work best with low level PCs advancing at least every 2-3 4 hour (tabletop) sessions, compared to higher level PCs advancing around every 4-6 4 hour sessions, perhaps slower at very high level.

For very high level PCs. 11th+, I don't recommend sticking to a metric by level, the 10th level award of 3000 XP per session serves as a reasonable minimum guide. That gives 5 sessions 11-12 (15000 XP), but at that rate would be 16-17 sessions to go from 19th to 20th.

Why 'Lost Mine of Phandelver' sucks

Inspired by common recommendations to use it as a GM's first 5e adventure, I have to say I disagree.

The problems start with the very first encounter, and what it says about the world.

Why are 4 goblins ambushing a well armed party of 4-6 larger and likely tougher looking travellers - the PCs?  Answer: To give PCs a starting encounter. From goblin perspective it makes no sense. They would let such go past & wait for easier prey.

Why do the goblins use such sub-optimal tactics, with 2 charging into melee to get slaughtered? Answer: So the PCs can win. Because if the goblins were to use sniping tactics from cover, combined with their racial ability to hide, disengage etc, they might well actually win.

This is a terrible first encounter which sends the message that the world exist for the benefit of the PCs. It would work much better to (for instance) have four drunken goblins lounging around the wagon Paizo-style, a wine butt spilling the last dregs onto the ground - ie the PCs got lucky. Then the goblins can even react in confused & suboptimal manner without straining plausibility.

But don't set up an ambush that is suicidal from the POV of the ambushers.

Edit: Discussion here

Improving Initiative

Sunday was my first chance to use my modified initiative approach in a live 5e D&D tabletop game - every player rolled init as usual vs a target number of enemy DEX mod +11, then winners could act in any reasonable order or as determined by GM (eg those at front going before those at back). After enemy's turn the losers went, along with the winners getting their 2nd turn, and so on.

On Monday I did the same in my White Star game, an sf system based on old school D&D. Here the target number is 4, initiative is rolled on a d6 plus PC DEX mod of typically -1, 0 or +1.

This worked really well in both cases and made combats fly past MUCH faster, so I'll definitely be sticking with it. I think the only game where I wouldn't use it is 4e D&D.

Mass Battle Stats - scaling things up

In most RPGs, the combat stats for an individual might just as well be the stats for a unit. Here are some thoughts on effective scaling:

The main thing is to choose an appropriate scale. Generally speaking, you want a 2-digit number of figures/chits on the board for it to feel like a battle, and something like 20-50 figures is ideal.

Times Ten Scale - "Battlesystem"
A unit is 10 men - a squad. A group of units is a company of ca 100 men.
Distances on the board/battlemat are x3, feet become yards, so 5' > 15' or 5 yards.
Time is also x3, so each unit round is 3 rounds of individual combat. Resolve 3 individual rounds for your PCs, then roll a unit round to see how the army is doing.

Times One Hundred Scale - "Hastings"
A unit is 100 men - a company or century. A group of units is a brigade of ca 1,000 men.
Distances on the board/battlemat are x10, so 5' > 50'.
Time is also x10, so each unit round is 10 rounds of individual combat. Resolve 10 individual rounds for your PCs, then roll a unit round to see how the army is doing.

Few battles will need a larger scale, but...

Times One Thousand Scale - "Cannae"
A unit is 1000 men - a brigade. A group of units is an army of ca 10,000 men.
Distances on the board/battlemat are x30, so 5' > 50 yards, 150'.
Time is also x30, so each unit round is 30 rounds of individual combat. Resolve up to 30 individual rounds for your PCs, then roll a unit round to see how the army is doing.

Examples:
A force of 200 men becomes 20 units, in 10-man squads.
A force of 2,000 men becomes 20 units, in 100 man companies.
A force of 20,000 men becomes 20 units, in 1000 man brigades.

Resolution
Treat each figure as a unit of the appropriate scale. Roll initiative etc as normal. One roll per side is fine, but individual rolls by organised group of units (company/brigade/army) or their leaders works well too, and may better reflect the disorganised state of battle.
Have groups of figures/units organised into cohorts/brigades of ca 10 act on their turns.
For a unit group, roll to hit for all units/figures. Use average weapon damage. Add up the number of hits and multiply by damage. Remove figures whose hp equal up to that damage total. Fractions can be ignored, since some damage is bound to be wasted, except for lone units.
Use morale rules as normal for your game.

Made it!

Today the pub where I host my Wilderlands D&D Meetup said there was Rugby on TV, so they wanted a £260 credit guarantee (bargained down from £300) to let us play there. I had been going to say no, but my players convinced me otherwise, that they could eat and drink that amount - and they did, £290! Made it! :D

This may have been our last week with only three GMs running, all Wilderlands, all same campaign world and map, just a few miles apart. Next week we're adding a 4th GM in a linked campaign setting, which will let us increase size to 28 (24 players & 4 GMs) - and one of my players is pretty keen to make it 5 GMs if I let him go... :) ...That'd let us have over 32 people playing 5e D&D in the Wilderlands of High Fantasy!

3e/PF to 5e Conversion

I ran a bunch of Pathfinder adventures using 5e. It is definitely harder than running OSR material in 5e, but remains doable. I had several handy monster conversion metrics:

Halve AC over 10. Halve stats over 20 (& cap at 30, obviously). Increase damage +50%. Increase hit points +50% at levels 1-10; at very high level the listed hp are often fine as-is. Ignore feats and weird crap on multi-page stat blocks, but look out for signature abiliities to convert.
Finally, calculate attack bonus & save DCs using the 5e rule of Proficiency + Stat bonus, +8 if it's a save DC.

This works great 95% of the time. I occasionally fluffed when eg converting to Legendary status, the headless horseman who could throw a 15d6 hellfire ball 4 times/round was a mistake - with massive damage spell-type attacks you may want to keep the 3e/PF damage as listed.

5e D&D Followers & Henchmen

I think the big lacuna in 5e DMG for domain/lord play is the lack of follower rules for high level PCs. So I've been adapting the 1e DMG rules, which work well - eg I rolled on the Fighter table for a Barbarian-14. I use the Thief guild rules for Rogues establishing networks. For a Cleric establishing an urban temple I used a formula for cult followers = (level + CHA bonus)x5 - but if she establishes a proper holy fortress I'd probably use the DMG armed followers table for her too.

For PC-class Henchmen I'm also using the 1e DMG rules, with a level limit of 1 per Tier, which closely matches the DMG limits. So:

PC Level > Henchie level
1-4 > 1
5-10 > 2
11-16 > 3
17-20 > 4
20+Boon > 5

Max number of henchmen = PC CHA bonus+2, so from eg 1 at CHA 8 to 7 at CHA 20.

Some early morning thoughts on 5e D&D

Been running 5e D&D both online & tabletop for just over three years now. A few thoughts.

The monster creation rules are garbage, I just have to do it 3e style and eyeball a CR.

The encounter-building rules are also garbage, I do it 1e style and ignore them, which works great in 5e.

The rapid advancement from 1st to 3rd is one of those 5e things that works much better in practice than on paper, and has significant potential gameplay benefits when combined with 'bounded accuracy'. Unlike 3e-4e or even really 1e-2e, I can start new players & new PCs of current players out at 1st level alongside 5th level veteran characters. They can contribute to the group, and if they survive they can catch up fast, hitting 3rd in a couple sessions. And there is something very special about starting D&D at 1st level. Partly the new PC is easy understand, partly there is kind of a freshness to them, their whole life ahead of them, combined with a feeling of vulnerability.

XP - Been playing around with this a lot. The 5e XP system is over-focused on monsters I think, and tends to make wandering monsters a source of welcome XP rather than a threat. I was giving 1/5 monster XP but been finding that too little, and tending to over-award "quest" XP to compensate. Currently going with 0.5 standard monster XP, plus XP for gold looted from the dungeon & returned home. Did that last session & seemed to work; the party didn't do great on finding treasure so total was about 3/4 what they'd have got from standard full monster XP. 11th+ PCs will get only 1/10 treasure XP though, since 11-20 treasure awards are much bigger while XP per level almost flatlines.

Advancement Rate:
Overall for my sandbox Wilderlands I generally want an advancement rate of around 3-5 sessions/level, from 1-3 sessions at 1st-4th going to maybe 5-8 sessions at levels 5-10 and much slower, around 10 sessions/level, at 11-20. This is generally what I'm seeing in play with the XP chart as written, though with occasional spikes - 2nd level PCs carry off the orc hoard, or high level PC solos the dragon/lich/BBEG. That's great, advancement spikes are a good thing in my book.

Getting Started With Sandboxing in Fantasy RPGs

The best advice I can give: 

i. Make a small starter sandbox using hexes, say 1 mile/hex, around 300-400 squares miles is good.
ii. Fill it with stuff - your ideas, published site based adventures, rolls on donjon.se et al, liberally plastered over the hexes.
iii. Have a PC-friendly starter village towards the middle, with a few named NPCs, and useful services such as an Inn, marketplace, smithy, and lord's hall.
iv. Give the PCs 2-3 starting rumours of nearby adventure sites.

Do it right and you have opened the door to many years of great adventures.

Death of High Level Characters

From http://www.therpgsite.com/showthread.php?38327-Death-of-high-level-characters

Should a GM let high level characters die?

I certainly would let them die, if that's the natural outcome of play. But old school D&D rules tend to make high level PCs robust, and also tend to make the players of high level PCs highly skilled! So it's not a common event. And it tends to be a dramatic, epochal event when it happens.

I think it was in 2016 I last saw a really high level PC perma-death: At the climax of my BECMI campaign, my son's PC the Duke-Sultan William of Castellan/Karameikos/Alasiya, an MU 18 (boosted to 19th caster via Ioun stone) was pressing home the attack against the Heldannic Knights during the Soderfjord invasion of Ostland. He had just nailed the Heldannic Knights-Marine with a well-placed fireball as their Warbird's bay doors opened, saving his party from being overrun by a couple dozen high level Fighters, when he was Disintegrated by the Heldannic Warbird's Bile Belcher cannon. AIR he was targetted 3 in 6 and would have survived on a roll of 4+ on d20, but made me roll and I got a 2. William was gone, after years of play (he had started at 4th around 4 years previously, when my son was only 5 years old!) but his side was victorious, King Hord Darkeye of Ostland was defeated, his kingdom conquered, and William was mourned by an assembly of high level heroes, including his son Bravery the new Duke-Sultan, the High Priestess Lady Roseanna and Baroness Alexandra Vorloi, all great champions in their own right.

In that case I was already planning to end the campaign with that session, although I did end up doing some epilogue stuff.

No very high level (15+) PCs have perma-died in my 5e Wilderlands sandbox game yet; High Priest Thuruar (Clr-15) did perish to the hellfire ball of the Death Knight Varek Tigerclaw*, but Hakeem the Barbarian defeated Varek and was able to get Thuruar raised by Lady Meda of Thusia. If a really major PC like Hakeem (now Barbarian-20) did die it would certainly alter the campaign (Hakeem's newly founded empire would almost certainly collapse, for a start) - but it certainly wouldn't end the campaign.

*A former PC reanimated by the Necromancer Borritt Crowfinger, BBEG of the campaign for many years until his final demise a few months ago.
___________

 Originally Posted by mAcular Chaotic 

Ah, so there's no main quest of sorts, so there's no big deal if one group dissolves because they were never going to stick around in the first place.

The open campaign is bigger than any one quest, but I've seen major (like 20+ session) quests emerge from play, like the time a group of PCs decided to quest far to the north of the normal campaign area to find and destroy the Gate Castle of the Black Sun. We still had a fair bit of PCs swapping in and out over those 20+ sessions, but I'd say there were 3 core PCs, and if all had died then that particular quest might logically have been abandoned/failed.

 Originally Posted by mAcular Chaotic 

What was his reaction to the death?

Some tears, but he took it manfully, especially for a nine year old.  Since his PC was almost an Archmage I accepted his request for a 'force ghost' scene where Sultan William appeared to his widow & mother of his 4 children, Princess-Sultana Adriana the daughter of King Stefan Karameikos, back in Karameikos where it had all begun 25 game years & 4 real years before. He bid her a very touching farewell & departed to the higher worlds.

If people come and go based on whatever quest their PC is interested in, what happens if your one PC is interested in something nobody else is? Does that mean you have to rely on coincidence to be able to join whatever session is happening? Like, I could imagine 20 different PCs, with like 12 different goals spread between them, making a fairly fractured group that may never actually get together in any real combination.

Well if the player is playing online or if I see them regularly offline (eg my son Bill) then I can do them solo sessions. Eg Bill has a plan for Shieldbiter his Barbarian-17 Dragonborn to solo the ancient black dragon Matriarx of Dread Isle, we'll do that at a suitable time. If the player can only make my regularly scheduled Sunday tabletop sessions then I would need to resolve any solo plans abstractly. If they don't want to take part in a particular quest the others want then they could sit that out, make a new PC for that quest etc.

Normally players & PCs are happy to team up for adventure, whoever's idea it is, as there are rewards in fun, gold, XP etc. But I do see stuff vetoed, eg the PC Hakeem refused Shieldbiter's request to help him go up against the ancient red dragons beneath Fortress Badabaskor, rumoured to possess the sacred Dragon Armour and Dragon Shield of ancient Arkhosia.

Edit: Rely on coincidence - yes just like in the comics (Savage Sword of Conan, notably) and pulps, heroes are always running into each other in unlikely circumstances.  But I also have a home base, Selatine port village, that serves as the heart of the campaign, the starting point for (most) new PCs & PC groups, etc. With lots of PCs based there they naturally interact and form adventurer parties.

Oh, also re the tabletop element, I would be willing to run up to two separate groups on alternate weeks if necessary, if the PCs split up. And I have a co-GM who helps out, running the Bratanis region, a low level campaign area west of Selatine. PCs can go back and forth from there, being GM'd by him one week then me the next.

 Originally Posted by Ravenswing 

partly due to what I call the Tasha Yar Rule, which boils down to that I'm not going to kill a PC for no better reason than a grunt orc made a good roll.

Mileage... I *love* it when some mook NPC gets lucky and takes down a high level PC or BBEG!  Not something I see often in 5e, but the Blight Belcher disintegrator cannon that took out legendary wizard Duke-Sultan William Karameikos was crewed by zero level nobodies... Who were then swiftly dispatched by the furious Sir Bravery, William's elder son, in a rage worthy of Achilles.

Bill also lost his level 8 Dragonborn Fighter Drakonok to a horde of mook orcs a few weeks ago - turns out that in 5e challenging an entire orc tribe to battle is not such a great idea. Mind you when Drakonok's companions (who had fled the orc horde summoned by Drakonok's roared challenge) came back across the scene an hour or so later, the breath-frozen, stabbed, and arrow-pierced orc bodies were piled high in the corridors of Stonehell.

 Originally Posted by rawma 

With very high level characters, they can form parties implicitly to some extent

IME very high level PCs in sandbox open game tend to get played less often; they tend to be focused on political shenanigans rather than adventuring, and the player will start to spend more time with their lower level PCs adventuring in lower level parties. Eg in my Wilderlands currently one player has PCs of levels 5, 9 & 20, another has 4, 5 and 17, a third has 5 & 13. A 4th or 5th level PC can adventure with third or eighth level PCs, an 8th level can adventure with 5th or 13th, but very high level characters tend to dominate too much, aren't challenged by most adventures, & often have better things to do.

 Originally Posted by rawma 

The permanent loss of one of those characters would just mean that that player doesn't play at the higher level table until another character reaches a high enough level.

One thing I learned running an open sandbox campaign was to limit PC starting level; I settled on 8th as the maximum starting level, being the high end of what feels like mid-level in 5e. It doesn't work letting everyone roll up new 20th level PCs just because one guy got a character to 20th. Characters need to feel organically part of the world, and keeping them to 8th initially ensures that.

 Originally Posted by mAcular Chaotic 

In those kinds of games, do the PCs ever bond with each other and have any roleplaying?

I find the more PCs are independent entities, not just part of The Party gestalt, the more roleplaying I see, and the more interesting interpersonal relationships form among the PCs.

 Originally Posted by Madprofessor 

Agreed! A one on one solo adventure is an awesome way to play. Doesn't happen often though.

I like doing online solo high level political stuff, and some PCs like Hakeem the Conanesque barbarian PC are great to GM solo. Mostly though for tabletop I greatly prefer GMing a group, ideally 3+ players; these days I tend to feel a bit awkward running solo games, and there is the lack of player-player interaction sparking new stuff.

 Originally Posted by mAcular Chaotic 

Actually, a related question: how do you guys feel about long term campaigns ending in a TPK?

This isn't "two sessions in and everybody wipes," but having played for years and made it to a final boss, and then getting destroyed.

Good, bad? Just play it out and let it end on a sad note? Give them a way out?

This is only a meaningful question in single-party campaigns, especially linear ones with one goal. My Wilderlands sandbox has lots of different PC groups and a TPK would not be an issue; TPK of a very high level group probably means a big victory for some bad guys (or good guys!) but the world goes on.

In a single-party campaign, the group wipes, I may use that setting again later. Again BBEG victory will change the campaign world but it likely is still useable. I certainly won't fudge to keep them alive. It makes the game meaningless if the PCs can't lose.

 Originally Posted by Just Another Snake Cult 

Four of the players in my decades-long D&D campaign voluntarily retired characters after they got them to a point where, I guess, they felt like they "Won", or at least had a nice stopping point (Two took over domains, two married NPCs). They exited the stage of their own volition and effectively started over with new low-level characters. This was their own idea.

Is this common? Uncommon? Do most players keep playing a PC until they absolutely can't anymore?

In multi-PC games it's very common to retire or semi-retire the high level ones. Eg in my Wilderlands Hakeem Godslayer defeated the Black Sun, reached 20th level, founded the Empire of Altanis-Nerath. He's now basically retired from play, the player has a 5th & a 9th level PC for regular adventuring.

In an Adventure Path type game everyone plays the same one PC to the end of the campaign.

 Originally Posted by AsenRG 

Look at it like this: the new PCs get to start as underdogs, in a very different place!
I'd actually like something like this to happen.

I've done it a couple times recently - one group TPKs, another group starts off in the post-TPK milieu - but in both cases it was a couple years later IRL with new players as well as new PCs. Both times it made for a stronger campaign IMO; the new players were never in any doubt that failure was an option. 

The two cases I recall:

1. Rise of the Runelords AP - PCs were wiped out by Nualia at the end of Book 1 ca level 4, resulting in the destruction of Sandpoint town, the intended home base for the campaign. Years later Sandpoint was a goblin-infested charred ruin. Eventually the Black Dogs mercenary company was hired by Titus Scarnetti, one of the few survivors of Sandpoint, to clear the ruins - this went on mostly in the background of the new campaign though the new PCs did interact with Titus; Quillax the Druidess even delivered his wife's baby and ensured the child was born without the taint of Lamashtu (his wife & daughter had been captured by Nualia's Lamashtu ("mother of monsters") cult but ransomed back, so there was a fear they could have been infected). 

2. My Wilderlands - catastrophic PC defeat at 10th level, in the final battle the lead PC Varek Tigerclaw fell in the doomed defence of Bisgen town against the Necromancers of the Black Sun; leading to the rise of the evil Empire of Neo-Nerath. Subsequent PCs spent years battling the Black Sun as it embarked on a genocidal campaign of conquest; at one point the new hero Hakeem Greywolf battled Varek Tigerclaw, now a Death Knight of the Black Sun - and ripped out his black diamond heart. 

 Originally Posted by joriandrake 

I really have to take time to read the whole blog of yours. Was Varek player controlled for the battle with Hakeem?

Varek at that stage was a soulless husk of a Death Knight controlled by Borritt Crowfinger, NPC Necromancer & Prince of the Black Sun. So GM-controlled.

 Originally Posted by mAcular Chaotic 

What happened with that first group that TPK'd?

In "Burnt Offferings" the encounter with Nualia in Thistletop as written is a death trap. There's a narrow 5' tunnel with a pit trap/portcullis, it's perfect for a tough fighter like her to cut through the PCs 1-1. I was using 1e ADnD; the group hit the pit trap/portcullis (one PC* got his foot severed!), then got attacked by the yeth hounds & Nualia while disrupted, and were slaughtered. When I played it myself (in Pathfinder) the GM would have TPK'd us too, if he hadn't fudged massively, letting us flee then come back and redo the encounter (with Nualia still just sitting there, yuck) now with optimised tactics. Second time round my Cleric used 'Command' to force Nualia to come to us out of the kill zone, where we could deal with her.

*An unfortunate Paladin.

 Originally Posted by mAcular Chaotic 

Well, what is an unsuitable death. Is getting killed by a random encounter unsuitable?

I rem once in the early '90s Mortis Deathlord, a ca 25th level 1e AD&D PC in service to demon prince Graz'zt, was trekking through the mountains along with two similar level NPCs, on their way to assassinate the enemy King of Trafalgis, a viking realm. I rolled up a random encounter with 4 adult red dragons, who proceeded to wipe out these epic villain characters. 

It was definitely bathetic - and AIR I did have the PC resurrected some years later to continue service to his dark master. But the old stories have loads of bathetic ends. Jack Vance is inspirational.  The PC Mortis was ultimately executed by Graz'zt when he (a) started to get cocky, beginning to see himself as Graz'zt's equal, and (b) accidentally violated a non-aggression pact Graz'zt had with Thrin, a lesser god lawful good PC.

Originally Posted by mAcular Chaotic 

Here's another example, and one that might come up in a session soon: an assassin targets a PC in their sleep. This is appropriate for the context of the situation, but still, this is one of those situations where a PC, if the dice go bad, could die before they even know what happened.

If this happens I'll generally do the rolls at-table so players can see what happens. Many D&D versions make it hard to take out a high level PC at table. 

I don't make assassins omniscient. I recall one time a Neo-Nerath assassin struck as the ca 10th level PCs were resting at a mage conclave on their way to close the Black Sun gate. With the PC group was an NPC (from The Licheway, I recall) who claimed to be an Archmage, he dressed the part but was really a phony, a 2 hd non-caster. Anyway, when the Black Sun assassin was choosing his target, naturally his top priority was to take out the enemy Archmage...  ...And the PCs slept peacefully all night.

I generally find player characters with powerful enemies do take precautions, and aside from 3e/Pathfinder I can usually have those precautions be effective, or dice for it (3e/PF is broken IME and often requires handwavium why high level caster NPCs don't use scry-and-die on the PCs). I recall one serious assassination attempt - the PCs were in Ahyf, local hive of scum and villainy, and got into a big fight at the fighting pits liberating some pit fighters. Hakeem was badly wounded killing Gorok Halfogre the pitmaster, giving three Black Sun assassins who had infiltrated the town a chance to strike when he & his friends fled town. It resulted in a brutal night battle on the road but the PCs managed to prevail and took shelter in a farmhouse, fearing more assassins.

If Hakeem or similar high-value PC was foolish enough to put themselves in a position where an assassin could strike as they slept, I would roll with it, using the 5e rules they would take a critical hit in damage, plus poison - likely would kill a Wizard PC or similar squishy, but a high level Barbarian or similar might well survive the wound, wake up and kill their attacker. It would take a lot of CR 8 Assassins to take out Hakeem, 20th level Barbarian with 2 Epic Boons inc Epic Fortitude, CON 24 & 325 hp (takes half damage from weapons while Raging, so effective 650!). My son Bill's Dragonborn Barb-17 Shieldbiter with 209 hp might be more vulnerable, but generally it's the high level casters who really need to worry.