Rarity of High Level Characters and NPC Level Demographics

[mrguymiah]

I'm working on a program that will create, stat, and put together a simple relationship network for every npc in a population. The goal being to know exactly how many smiths are in any given town, which ones are capable of making that item the players want, etc.

And this brought up a distinct problem; how rare are higher level npc's in a given population? Most settings tend to assume there's a level 13 cleric kicking around to rez PC's, but that could easily be distortion based on having that need in most settings.

I figure I'll already need mutliple settings for the program to chart different playstyles, and here's what I've compiled so far, by looking at various threads:

"NPC's are Fodder": Lvl 1 NPC's make up 99% of the population. Anyone who isn't important is a 1hp mook.
Average Level: Lvl 3 NPC's in general. Lower levels being apprentices. Higher being those more exceptional people who make up civic leaders and royal guards. This assumes some level of growth over an NPC's life, even if it's not to the extraordinary rate of PC's.
"We're all Heroes Now": Lvl 6 NPC's in general. Fantasy worlds are a bit more difficult to live in; bugbears, orcs, goblins, and any other number of fantasy dangers mean even the average peasant gets some levels throughout their lives.

[Eldariel]

I'm working on a program that will create, stat, and put together a simple relationship network for every npc in a population. The goal being to know exactly how many smiths are in any given town, which ones are capable of making that item the players want, etc.

And this brought up a distinct problem; how rare are higher level npc's in a given population? Most settings tend to assume there's a level 13 cleric kicking around to rez PC's, but that could easily be distortion based on having that need in most settings.

I figure I'll already need mutliple settings for the program to chart different playstyles, and here's what I've compiled so far, by looking at various threads:

"NPC's are Fodder": Lvl 1 NPC's make up 99% of the population
Average Level: Lvl 3 NPC's in general. Lower levels being apprentices. Higher being those more exceptional people who make up civic leaders and royal guards.
"We're all Heroes Now": Lvl 6 NPC's in general. Fantasy worlds are a bit more difficult to live in; bugbears, orcs, goblins, and any other number of fantasy dangers mean even the average peasant gets some levels throughout their lives.


DMG has its charts. Pages 137-140 describe how you roll the highest level character for each inhabitation and then use that to generate lower level characters as appropriate. You can use those for a "high average"; by those rules, you'll end up with a significant number of very high levels. Metropolis can have up to level 28 commoners (which are rounded down to 20), though stuff like Druids and Clerics is capped at 18 and without modifiers, Wizards and Sorcerers at 16 (and Pallies and Rangers at 15 if you care). And those will generate a significant number of middle/lower level stuff - you take the highest level ones and their number and halve their level and double the number, and repeat. Thus you get a lot of chaff. Each inhabitation of 25001+ creatures is a metropolis with 4 rolls at 12+Class Modifier for each class. Such a world is quite high powered though.

Eberron, E6/8, etc. require quite different approaches altogether, of course. This might work for FR or Greyhawk which have stupid numbers of high level guys. Also, low size locations can generate ridiculously high level Druids or Rangers by the rules, since Thorps and Hamlets roll at +7/+8 instead of -3/-2 for those two.

[mrguymiah]

DMG has its charts. Pages 137-140 describe how you roll the highest level character for each inhabitation and then use that to generate lower level characters as appropriate. You can use those for a "high average"; by those rules, you'll end up with a significant number of very high levels. Metropolis can have up to level 28 commoners (which are rounded down to 20), though stuff like Druids and Clerics is capped at 18 and without modifiers, Wizards and Sorcerers at 16 (and Pallies and Rangers at 15 if you care). And those will generate a significant number of middle/lower level stuff - you take the highest level ones and their number and halve their level and double the number, and repeat. Thus you get a lot of chaff. Each inhabitation of 25001+ creatures is a metropolis with 4 rolls at 12+Class Modifier for each class. Such a world is quite high powered though.

So assuming I use that as the base for the "We're All Heroes Now" sort of option, what do you think for the more middle of the road option? Maybe hard cap at level 10?

[ngilop]

my demographics are kind of complicated


For humans, halflings and other short lived and kind races 30% are 1st level commoners 20% are 1st level NPC classes other than commoner 20% are level 2-4 npc classes 15% are 1st level PC classes and the rest are 2nd and above PC classes

Dwarves, elves and other long lived and kind races are different
15% are 1st levl commoners 15% are 1st level NPC classes 30% are level 2-4 NPC classes 15% are level 1 PC classes and 10% are level 2-4 PC classes and the rest are 5th and above PC classes.

[Sagetim]

I mean, there's also magic's effect on society coming in as a factor. After all, if you have enough increases to effectiveness due to some wandering druids and clerics dropping Plant Growth and other things to help farmers out, then you don't need the historic assumption of 90% of people being under/uneducated farming commoners. So one thing you might want to put into your program is the ability to adjust that assumption's percentage.

From there, you can use the rolling charts in the DMG for the guts of this operation. Add in some spaces to let people throw in their own modifiers for custom campaign shenanigans, and you're probably good to go. After all, if someone is running a game where a Benevolent Wizard King has successfully mandated public magical education for all of his subjects, or has bound the ghosts of mages past to make them haunt the dreams of children and teach them magical lessons in the unescapable classrooms of their dreams, or whatever, and give a campaign setting wide modifier for generating more arcane classes...or even outright replacing commoners with Magewrights from Ebberon.

Also, while the 'unimportant commoners have average stats in the 10's and 11's' is a simple shorthand for statting up npcs that you frankly don't care to put much effort into, it's not the most realistic thing to do. If you're working hard in on a farm all day, you're probably going to wind up with an above average strength and constitution, if nothing else. But that's sidetracking, and might be a better subject for a thread on it's own, something like 'what would representative stats look like for commoners of various trades?'.

[mrguymiah]

I mean, there's also magic's effect on society coming in as a factor. After all, if you have enough increases to effectiveness due to some wandering druids and clerics dropping Plant Growth and other things to help farmers out, then you don't need the historic assumption of 90% of people being under/uneducated farming commoners. So one thing you might want to put into your program is the ability to adjust that assumption's percentage.

From there, you can use the rolling charts in the DMG for the guts of this operation. Add in some spaces to let people throw in their own modifiers for custom campaign shenanigans, and you're probably good to go. After all, if someone is running a game where a Benevolent Wizard King has successfully mandated public magical education for all of his subjects, or has bound the ghosts of mages past to make them haunt the dreams of children and teach them magical lessons in the unescapable classrooms of their dreams, or whatever, and give a campaign setting wide modifier for generating more arcane classes...or even outright replacing commoners with Magewrights from Ebberon.

Also, while the 'unimportant commoners have average stats in the 10's and 11's' is a simple shorthand for statting up npcs that you frankly don't care to put much effort into, it's not the most realistic thing to do. If you're working hard in on a farm all day, you're probably going to wind up with an above average strength and constitution, if nothing else. But that's sidetracking, and might be a better subject for a thread on it's own, something like 'what would representative stats look like for commoners of various trades?'.

I will admit, I'm not so far as to generating attribute scores, yet. I was going to have it do the 3d6 method, since that's the most basic and doesn't give the higher 'heroic' numbers. From there, my plan was to try and slot people into jobs according to those attributes.

As for a variable urbanization value, I'm concerned that may be out of my league to determine the precise numeric effects that increased yield production would have at the wider scale. I imagine there are entire College Thesis' dedicated to the subject.

If you'd like some insight into how I'm working this massive project of statting entire populations; I'm coming at it from two directions. Utilizing the guidelines of "Medieval Demographics Made Easy" (http://www222.pair.com/sjohn/blueroom/demog.htm), I'm determining approximately how many urban, specialized profession positions should be available. On the other end, I'm generating a randomized population. My idea from there is to try and match people to the jobs based on several factors (level, ability scores, and race). So if there are 100 guard positions open, it should prioritize people with strength and constitution. If 10 of those are royal guards, it should prioritize the highest Ability Score and highest level characters to those positions.

In no way should this be a definitive for a DM, if they use it, but the idea is to give them an easier time answering questions like "how many smiths are in town?", "Is there a wizard who can cast X spell for me?", etc.

[mrguymiah]

Eureka! I figured out why I was having such trouble with my calculations. I was doing the math linearly and trying to create a split. The issue being that it was basically impossible for a young person to have more levels than someone older than them. I was simply conflating life experience with level gain.

However!

I then remembered something Doc_Maynot was telling me about from an L5R campaign; Exploding dice. Those who sought adversity once, and gained a level from it, are more likely to make a similar gamble in the future! (And that can represent a merchant making a risky business decision or a smith pushing himself to the edge of his skill, not merely combat challenges.) And thus, by utilizing an exploding dice recursive method, characters who roll level growth in the first place, will gain more levels, properly simulating the divergence of the steady growth population and the ambitious persons who seek challenge.

Here is the breakdown from my latest test, utilizing a population of 450k (Approx the population of Brevoy in the PF standard Golarion setting). It seems pretty accurate ot a "We're all Heroes Now" sort of setting like Golarion. Maybe just a _bit_ high, but I can adjust the numbers a bit and run it again.

Total - 450000
Lvl Array
Lvl 01: 76901 17.09%
Lvl 02: 95778 21.28%
Lvl 03: 91460 20.32%
Lvl 04: 72542 16.12%
Lvl 05: 49859 11.08%
Lvl 06: 30747 06.83%
Lvl 07: 16989 03.78%
Lvl 08: 8677 01.93%
Lvl 09: 4082 00.91%
Lvl 10: 1788 00.40%
Lvl 11: 739 00.16%
Lvl 12: 286 00.06%
Lvl 13: 98 00.02%
Lvl 14: 41 00.01%
Lvl 15: 9 00.00%
Lvl 16: 3 00.00%
Lvl 17: 0 00.00%
Lvl 18: 1 00.00%
Lvl 19: 0 00.00%
Lvl 20: 0 00.00%

Average Level by Age
Age 20 30 42 56 76
Level 1.86 2.57 3.17 3.66 4.05

Average Age: 64.47
Average Level: 3.39
Highest Lvl Char: 18 74
Lowest Lvl Char: 1 109 (Poor dude got NO successful dice.)

[Sagetim]

Nice breakthrough. My main question would be, following that distribution of levels, what kind of impact do you think that has on the productivity and economy of the kingdom in question? This wouldn't need to necessarily play in as a hard coded part of your program, but rather act as a guideline for making adjustments in job distributions. After all, if you have a significant number of mid level farmers, they're going to out produce the typical assumption of the majority of the population being level 1 commoners and having to be farmers just to make enough food to feed the kingdom, allowing for more experts, more educated people, more adventurers, and what have you. And that's not even mentioning how an abundance of mid level spell casters can drastically impact a kingdom's economy.

Something else to consider would be how many people might pursue greater education after having secured a successful income for themselves by other means. After all, an ambitious commoner might still manage to get some wizard training if they can get enough cash together from being really really damn good at farming. So instead of seeing npcs who are just 'Commoner 10' you could easily wind up with 'commoner 6/wizard 4' or 'commoner 3/Fighter 7' or even people who felt a calling in life to service in their religion and become 'commoner 4/cleric 6'. If you could figure out some kind of ambitiousness modifier for the people who keep taking risks to gain levels faster, with it being an adjustable value so that DM's can say 'no no, people are much less/much more ambitious in my setting', it could be interesting if nothing else.

As for ability scores, rolling random values is quick and dirty, but in real life a lot of your capability comes from what you do consistently. Sure, there's natural limitations or leanings in there, but, for example, if you sit around all day eating and drinking and partying, then you're just not going to have as high strength and constitution as someone who works the fields or does other regular manual labor/training. So I would think stat minimums or modifiers based on occupation might help represent that. Even given that different people are going to approach problems in different ways, that seems more like something you can apply when it comes to specific npcs instead of the average joe farmer and john smith.

[mrguymiah]

Nice breakthrough. My main question would be, following that distribution of levels, what kind of impact do you think that has on the productivity and economy of the kingdom in question? This wouldn't need to necessarily play in as a hard coded part of your program, but rather act as a guideline for making adjustments in job distributions. After all, if you have a significant number of mid level farmers, they're going to out produce the typical assumption of the majority of the population being level 1 commoners and having to be farmers just to make enough food to feed the kingdom, allowing for more experts, more educated people, more adventurers, and what have you. And that's not even mentioning how an abundance of mid level spell casters can drastically impact a kingdom's economy.

Something else to consider would be how many people might pursue greater education after having secured a successful income for themselves by other means. After all, an ambitious commoner might still manage to get some wizard training if they can get enough cash together from being really really damn good at farming. So instead of seeing npcs who are just 'Commoner 10' you could easily wind up with 'commoner 6/wizard 4' or 'commoner 3/Fighter 7' or even people who felt a calling in life to service in their religion and become 'commoner 4/cleric 6'. If you could figure out some kind of ambitiousness modifier for the people who keep taking risks to gain levels faster, with it being an adjustable value so that DM's can say 'no no, people are much less/much more ambitious in my setting', it could be interesting if nothing else.

As for ability scores, rolling random values is quick and dirty, but in real life a lot of your capability comes from what you do consistently. Sure, there's natural limitations or leanings in there, but, for example, if you sit around all day eating and drinking and partying, then you're just not going to have as high strength and constitution as someone who works the fields or does other regular manual labor/training. So I would think stat minimums or modifiers based on occupation might help represent that. Even given that different people are going to approach problems in different ways, that seems more like something you can apply when it comes to specific npcs instead of the average joe farmer and john smith.

Remember, this particular distribution was a test for the entire KINGDOM of Brevoy. So those higher level characters are guards, knights, nobles, etc. Most of those lower level characters are farmers, craftsmen, etc. I think this also answers your second point. Maybe once I've got a more fully functional model I'll include more specific code to handle the exceptions to the rules, but I think the current model handles things relatively well.

While you're right, I was planning to do the ability score relationship the other way. By assigning AS values before jobs, I can try and apply jobs based on their scores. That being said, I agree that this will create situations that may seem unrealistic at first glance, but I'm relying on the law of averages here. Since some of these characters will, through the random chance, be better suited to their work than others, their increased production is balanced by farmers with lowered production. Just because they're not suited for it means they're free from the feudal contract.

The next step in the program is to have the age roll be weighted towards younger people, as the current numbers are skewed, considering most medieval people didn't live to their max age.

[Seerow]

I'm working on a program that will create, stat, and put together a simple relationship network for every npc in a population. The goal being to know exactly how many smiths are in any given town, which ones are capable of making that item the players want, etc.

And this brought up a distinct problem; how rare are higher level npc's in a given population? Most settings tend to assume there's a level 13 cleric kicking around to rez PC's, but that could easily be distortion based on having that need in most settings.

I figure I'll already need mutliple settings for the program to chart different playstyles, and here's what I've compiled so far, by looking at various threads:

"NPC's are Fodder": Lvl 1 NPC's make up 99% of the population. Anyone who isn't important is a 1hp mook.
Average Level: Lvl 3 NPC's in general. Lower levels being apprentices. Higher being those more exceptional people who make up civic leaders and royal guards. This assumes some level of growth over an NPC's life, even if it's not to the extraordinary rate of PC's.
"We're all Heroes Now": Lvl 6 NPC's in general. Fantasy worlds are a bit more difficult to live in; bugbears, orcs, goblins, and any other number of fantasy dangers mean even the average peasant gets some levels throughout their lives.



The way I do it is this:
-Baseline assumption for adult characters is 3hd in an NPC class. So a Commoner3 is about the weakest thing you're going to run into.
-Around 30% of the population is made up of NPC classes other than commoners (eg Aristocrat, Expert, Warrior, etc).
-Around 10% of NPC classed characters are level 4 or higher. For each level beyond 4, you have half as many.

-PC classes make up approximately 5% of the population.
-PC classes start as NPC3/PC1. They alternate between gaining a PC level and trading out an NPC level, until all NPC levels are gone at level 6.
-These follow the same rules as NPCs (so 10% to level 2, then half thereafter).


So starting off with a large city of say 10,000 population.

You've got approximately 6,500 commoners. Of these around 650 are level 4. 325 are level 5. 162 are level 6. 81 are level 7. 40 are level 8. 20 are level 9.

You have approximately 3,000 NPC classed characters. Of these, around 300 are level 4. 150 are level 5. 75 are level 6. 32 are level 7. 16 are level 8. 8 are level 9.

You have approximately 500 PC classed characters. Of these, 50 are NPC2/PC2. 25 are NPC2/PC3. 12 are NPC1/PC4, 6 are NPC1/PC5, 3 are PC6. And there is 1 7th level PC classed character in the town.

Of course these are all distributed among the different classes. So you have 8 9th level NPCs, that might be 2 Aristocrats, 2 Experts, 1 Adept, and 3 Warriors, or something along those lines.

[Yahzi]

I'm working on a program that will create, stat, and put together a simple relationship network for every npc in a population. The goal being to know exactly how many smiths are in any given town, which ones are capable of making that item the players want, etc.
I wrote a program (http://www.drivethrurpg.com/product/217951/Sandbox-World-Generator)to build kingdoms. It doesn't go quite into that detail, but it does decide what Master craftsmen are in a town. (Do you really need more detail than that? I mean, does anyone want anything other than masterworks?) You might also be interested in Merchants of Prime and Lords of Prime, which talk about the assumptions behind the program.


And this brought up a distinct problem; how rare are higher level npc's in a given population? Most settings tend to assume there's a level 13 cleric kicking around to rez PC's, but that could easily be distortion based on having that need in most settings.
Most settings assume a profligacy of magic that would destroy the setting as presented. I went with option #1.

[unseenmage]

The book Power of Faerun has the demographics for noble and royal courts if that helps.

Once worked on a project using the numbers therein to build an all humanoid base race Effigy court in the wilds of Faerun.

[mrguymiah]

*snip*
My only gripe with that sort of method is that it feels a bit odd that there are that many high level characters, but that they're also all the SAME level. It's certainly not a bad system, though. Just concerned of its results at bigger populations numbers, though. It also pigeonholes all higher level characters into living in the bigger populated cities. Where as a more randomly generated system allows a high level charact to appear in a smaller settlement.


*snip*
I'll certainly check out the program! It looks interesting.

As for the options, I plan to have that be a toggleable setting. So that people who want their PC's to feel like goddarn BEASTS compared to the NPC's can do that. Those who want a more Ptolus-like feel to it, where there's a lot of high level NPC's can do that. I'm just trying to get a feel for what sort of rarities people tend to think of for their high level NPC's.


*snip*
I'll have to look into that, but I admit that I've heard more than a couple of people rag on Faerun. So I'm not sure I'd want to use it as a baseline, if it's as poor as they say.

[Yahzi]

I'm just trying to get a feel for what sort of rarities people tend to think of for their high level NPC's.
I get the impression that most people use a sliding scale; that is, there are more high levels when the PCs become high level. Certainly the design philosophy in 5E has been to absolutely obscure any kind of demographic data, which is odd given that apparently a handful of mooks can actually be dangerous in 5E.

I know the charts in the 2E DMG would have you walking through town and just randomly bumping into a 14th level illusionist. Like those guys just grew on trees or something.

Simulation has always taken a back seat to storytelling with D&D. Which is a problem for those of use who use simulation to tell stories. :smallannoyed:

[mrguymiah]

I get the impression that most people use a sliding scale; that is, there are more high levels when the PCs become high level. Certainly the design philosophy in 5E has been to absolutely obscure any kind of demographic data, which is odd given that apparently a handful of mooks can actually be dangerous in 5E.

I know the charts in the 2E DMG would have you walking through town and just randomly bumping into a 14th level illusionist. Like those guys just grew on trees or something.

Simulation has always taken a back seat to storytelling with D&D. Which is a problem for those of use who use simulation to tell stories. :smallannoyed:

You're telling me! Ran into another thing where I debate realism vs game mechanics. Ability score assignment and the fact that ALL professions are based on Wisdom. Thinking I'll focus realism, rather than straight game mechanic.

[Yahzi]

You're telling me! Ran into another thing where I debate realism vs game mechanics. Ability score assignment and the fact that ALL professions are based on Wisdom. Thinking I'll focus realism, rather than straight game mechanic.
All professions are based on WIS? Huh, I guess I houseruled that away so long ago I forgot it was ever true.

One thing I had to do was change the XP curve to double at every level. Otherwise even quite small kingdoms produce very high rank people. I can justify a pseudo-medieval world up to 9th rank without too much trouble... but it gets increasingly dodgy with every level higher.