Fixing the DMG Demographics

[seasong]

The advancement rates I gave assume that level 1 characters come upon a CR 1 threat in groups of 4 (yes, I divided the XP ). People don't face threats singly if they can help it - they're as bad as kobolds that way.

By the book, a CR 1 encounter costs a 4-person party of level 1 characters 25% of their resources (in hit points, potions, spells). On rare occasions, that kills someone.

 

[S'mon]

The advancement rates I gave assume that level 1 characters come upon a CR 1 threat in groups of 4 (yes, I divided the XP ). People don't face threats singly if they can help it - they're as bad as kobolds that way.

By the book, a CR 1 encounter costs a 4-person party of level 1 characters 25% of their resources (in hit points, potions, spells). On rare occasions, that kills someone.

If they're Commoners it kills the lot of them (them being CR 1/4) 50% of the time, if they don't flee first...

Of course you can give Commoners as much xp for a CR1/4 encounter as a PC gets for a CR1 encounter, if you like. Would certainly help explain 20th level Commoners.

Likewise, my '50% advancement' rule assumes that you don't always meet a CR-appropriate challenge, if you're an NPC - in fact, most challenges are probably either trivial (few xp) or fatal (death); it's the few that lie in-between, the ones that seek out PCs like homing missiles, that enable advancement.

 

[seasong]

If they're Commoners it kills the lot of them (them being CR 1/4) 50% of the time, if they don't flee first...

Of course you can give Commoners as much xp for a CR1/4 encounter as a PC gets for a CR1 encounter, if you like. Would certainly help explain 20th level Commoners.

Actually, if you take a quick look at the spreadsheet I included, I used lower CRs for commoners, and I assumed greater numbers of them. They advance considerably slower.

I ended up with this level comparison for commoners, in a kingdom of 1 million people:
Level - Population
0 - 545,772.2 (age 11 or younger)
1 - 82,732.4
2 - 67,013.2
3 - 77,421.4
4 - 71,622.9
5 - 46,991.7
6 - 36,713.4
7 - 24,804.9
8 - 13,182.3
9 - 7,800.3
10 - 3,350.7
11 - 2,607.7
12 - 919.7
13 - 779.3
14 - 461.7
15 - 369.5
16 - 31.4

 

[S'mon]

My approach, given a million Commoners:

Roughly 40% are children, 20% aged & infirm, all being noncombatant. 40% are fit, healthy Commoners: 400,000 Commoners Level 1 or higher. They're CR 1/4, so where Warriors & other CR 1/2 types have 25% levels 2+, Commoners will have 1/8, 50,000, level 2+, almost all exceptional individuals ("That which does not kill me..."), often in high risk professions like miner or prostitute, where Challenges are likely to be encountered.
That gives:


350,000 level 1
25,000 Level 2
12,500 Level 3
6,250 Level 4
3,125 Level 5
1,612 Level 6
806 Level 7
403 Level 8
202 Level 9
202 Level 10+

This is somewhat theoretical, in practice I've never had to do this for Commoners and might well go with 1/4 advancement at higher levels rather than 1/2.

 

[seasong]

The only problem I can see with that is that, at least in my experience, a total skill bonus 5-6 is needed to be reasonably competent, and a total skill bonus 10+ is needed to be "professional" in terms of how DCs work.

A professional farmer should also, with a team of 4-5 of his buddies, be capable of killing a few wolves (with losses). That also points to a commoner level-3 for any real competency.

Your statistics end up with very few at that level.

On the other hand, you could nudge those numbers just a bit, to balance the number at level-3 (very little death rate below 3), and then start at the 1/4th rate above that. Something like:

339,368 level 1
298,644 level 2
271,495 level 3
67,874 level 4
16,968 level 5
4,242 level 6
1,061 level 7
265 level 8
66 level 9
17 level 10+

 

[Xeriar]

I don't think advancement-by-age suits standard D&D at all - to my mind, and from reading the DMG's XP section, a CR 1 encounter is what defeats (and quite possibly kills) a single elite level-1 PC-class character 50% of the time, so high-levellers ought to be rare, and high level Commoners nonexistent.

Advancement-by-age is fine for variant settings like Traveller20, of course.

People face challenges all throughout life, of many different kinds, and learn from them. For my program, the encounter is spread out over an age category - 5 years for humans, on top of which it can be adjusted for a number of factors afterwords anyway.

 

[GuardianLurker]

Xeriar, Seasong:

Thanks muchly. I'm going to be twiddling lots of bits tonight I'm sure. (*Muwhaha* - tremble in fear little bits!)

As for Cr-based/age-based enhancement:
I think that's the best idea so far, although I think the curve there isn't quite so steep.

Maybe a rate of 1 level/2 years for levels 1-3, 1-2 levels/3 years for 4-9, and then a sudden drop off as the common CRs (<1-3) start dropping off the XP chart, followed by an increasing decline.

Also, I like the age-based stuff (in concept, since I haven't looked at it yet), but I worry at the divisions - the default age categories seem *way* to broad, and 5 year chunks just semi-arbitrary.

Also, something to keep in mind that if you want "wacked-out elves" (and similar long-lifers), instead of varying the years-per-age-stripe (which is what Xeriar seems to imply), you just keep it constant.

The problem is, without some form of "relative period" adjustment, no matter what advancement basis is used - CR, XPs/year, or whatever - the long-lived races (elves, dwarves, dragons, etc) will have *incredible* level maxes. And see the recent "who rules the world" thread for the problems that implies.

And now for the comments and queries:
Seasong - one of the issues this is trying to address is having the *need* to limit top city sizes; to provide the tools that allow the GM to see "if you want top city size X, you need _this_ kind of base (in magic and pop), and you'll get Y level adventurers."

One of the things I think we all need to avoid is the unstated assumptions, we're working with - like Xeriar's whacked elves, some will want that, so while we can include it (or not), we need to be prepared to defend/state why we did, and understand how it changes things when we do the other. Even just something as simple as changing the "base ratio" can give wildly different scenarios.

Also, I looked at the Census data, and at first glance, it seems to imply a slightly different grouping; in particular, instead of single cities, there seem to be "clusters". You have the top-most city, then 1-2 a significant fraction (75-80% of top) below it, then 2-3 the next step down, and so on.

Ultimately, the goal is to produce something very like a "Society Builders' Handbook" maybe even with tools, suitable for use in building societies for any campaign, not just psuedo-medieval-european ones. Heck , if we do it right, we might even be able to sell it to a publisher (or maybe not). Though if we intend to go that route, we 'll have to leave the boards soon.

 

[Xeriar]

One of the things I think we all need to avoid is the unstated assumptions, we're working with - like Xeriar's whacked elves, some will want that, so while we can include it (or not), we need to be prepared to defend/state why we did, and understand how it changes things when we do the other. Even just something as simple as changing the "base ratio" can give wildly different scenarios.

Well, for my program you can certainly adjust the CR base of a race to take such things into account, but... Seasong's method makes elves consistently 10-20 levels higher before perishing (if at such levels they're gonna perish at all)

Edit: To clarify, I don't want whacked out elves, and I don't think everyone does, either, so Seasong's method simply isn't for me.

 

[Xeriar]

Ultimately, the goal is to produce something very like a "Society Builders' Handbook" maybe even with tools, suitable for use in building societies for any campaign, not just psuedo-medieval-european ones. Heck , if we do it right, we might even be able to sell it to a publisher (or maybe not). Though if we intend to go that route, we 'll have to leave the boards soon.

For its attributes, d20 just breaks down with obscenely large numbers of people. Coruscant, for example, should have 25 people with all 17s or better. I don't believe for a minute that Yoda only started with a Wisdom of 15.

Maybe for d20 modern though, but it's still stretching it.

 

[Kid Charlemagne]

This is a topic that I've given some thought to over the years, and am still modifying my views... One thing I haven't seen anyone mention is the "law of diminishing returns" effect of CR's; namely that as you go up in level you get less XP for common challenges.

A Commoner may get XP for surviving a tough winter or bringing in a good crop, but after a certain point, that should cease to give any further reward, XP-wise.

In other words, a 1st level commoner may get decent XP for his first couple years on his own, but after 10 or so years unless there is a real good reason, he'll just stop learning new stuff. He won't have the opportunity to deal with a CR-4 kind of encounter, possibly in his entire life. That will limit the top end of commoner achievement considerably, I think.