Deity creation, making baby gods

[wlmartin]

Morning all!

I am working on building my campaign from scratch, including deities and their portfolio.
Part of this is going to be the pantheon.

This is mainly for Pathfinder but could apply to any D&D system

My design is 1 neutral "parent" god, 4 gods for each primary alignment (LN, CN, NG, NE) and then have them interbreed as they do, in the early years.
Since these beings are very primal, their alignment is going to be core to who they are, and how this is passed on is almost going to be genetic for my purposes.

So, I intend to create a formula to calculate how their alignments effect from their interbreeding and them base my story design around that.
I intend to use a bit of randomness and this is how I felt it was best

Run the following against each alignment type (once for the 1st character, ones for the 2nd character, aside from TN which will be handled differently)

#1 If alignment of both parties is identical, 80% chance of offspring sharing it, 15% chance of it being neutral, 5% chance of being opposed
#2 If alignment of both parties is opposed (Lawful vs Chaotic, Good vs Evil), 50% chance of offspring being neutral, 25% for either opposed alignment
#3 If one alignment is neutral, there is a 45% chance of either Neutral or the other alignment, 10% chance of being opposed to the other alignment
#4 If however they are both neutral, 80% chance of both being neutral, 20% chance of a random alignment other than true neutral (out of the remaining 8)

In the case of chosing the percentile position for which alignment (choosing which alignment goes first is alphabetical so in #2, 50% chance of neutral, 25% chance of one alignment, 25% chance of the other, you assume 1-50 is neutral, evil gets 51-75, good gets 76-100)

I think this puts a weight in dominant alignments (neutral really being more of a middle ground so less weight given to that in some cases) and using the above here are some examples

LG with NE, roll 56 and 83 for my randoms
1st is L vs N, #3 applies, it would be neutral
2nd is G vs E, #2 applies, it would be evil
--> so LG with NE

Lets try again this time with 6 and 32 as the rolls
1st is L vs N, #3 applies, it would be lawful
2nd is G vs E, #2 applies, it would be neutral
--> so LG with NE becomes LN


EDIT : I appreciate in this example its a 2nd tier (NE) and 3rd tier (LG) god interbreeding and that will be part of the process, there will be some intergenerational action going on to mix it up

This does more than compare the position on the 9-grid of alignments and adjust the step by 1 or something, this breaks down the component 2 elements and throws them together.

Is this realistic in PCs or NPCs... possibly not since they are a lot more complicated
Plus, its used to help me quickly come up with a splattering of raw alignments which i can fluff up... instead of having to come up with the story first and it taking x10 as long



I would welcome any thoughts on the formula itself as well as the concept
Any suggestions would be welcomed, especially ways to make it simpler (if possible)

 

[delericho]

Both the formula and the concept seem fine (though it's not what I'd do).

In terms of simplification, you might find it beneficial to write up the twelve tables showing the parent's alignments and the resulting child's alignment - that way you're just doing a table lookup followed by a single roll, rather than following several steps. That doesn't simplify the formula any, but it does simplify the procedure.

(The "twelve tables" are six for the possible L/N/C alignment combinations and six for the possible G/N/E alignment combinations. That is Lawful&Lawful, Lawful&Neutral, Lawful&Chaotic, Neutral&Neutral, Neutral&Chaotic, and Chaotic&Chaotic; Good&Good, Good&Neutral, Good&Evil, Neutral&Neutral, Neutral&Evil, Evil&Evil.)

 

[wlmartin]

Yes, that makes sense

With only a fairly small finite number of combinations it makes more sense to run it via a table rather than a series of conditional statements

So would you say that for any given alignment combination there is a certain chance and go from there

I'd for those that are the same there is a 80% chance at same, 15% neutral and 5% opposed so Alignment combinations are the rows of my table and the 3 possible options are in columns alongside their chance, repeated for both characters

My example below is for LE and LG
So in row A, col a we have LL, col b is L, col c is 80, col d is n, col e is 95, col f is C, col e is 100

So I run a single d100, compare the number against col c, e, f sequentially until the number found and then return the value against the column to its left

However in row b, I have EG, so my return columns are N, E, G and their probabilities are 50, 75, 100

In both cases if the result is 66, 92 for my 2 rolls
In part 1, with 66 its under 80 so the first return column hits, and L is the return
In part 2, it hits the 2nd result and gives me G since its between 75 and 100, my last value

So LE and LG create LG in this case
Using my formula there is a stronger weight to Lawful in this offspring but it's adore likely to get neutral or failing that one from either


Have I got your idea right?

 

[delericho]

I would actually deal with each of the two axes in a separate table, and use two lookups (one for each part of the alignment) just to reduce the complexity of the tables.

But otherwise, I think you've got what I'm saying.

 

[Celebrim]

So, at the end of this, you are going to end up with a list of alignments. As much as I'm a fan of the alignment system, this strikes me as a misuse of it, because the alignment system is only 9 buckets and what's interesting isn't the buckets themselves but what you choose to put in them. You are acting as if the buckets themselves are meaningful, when really they are just labels on a bucket and are defined by what's in them.

Rather than running the simulation as described, I suggest you back up and run the simulation with some initial abstract concepts. That way, the interbreeding of the abstract concepts will be suggestive of some new concept. Once you have the list of deities and their associated concept, you can place them in the bucket you feel is most appropriate.

I have used a system similar to the one you describe before, but not for the purpose of world building. Instead, I used it to determine the alignment of the children of PCs. However, the system was designed to also produce 'mutants' - whose alignment didn't match that of either parent - in a small percentage of cases. The list I used was something like: alignment of father, alignment of mother, one step more lawful than father (or lawful), one step more lawful than mother (or lawful), one step more good than father (or good), etc., ....lawful good, lawful neutral, etc. Each possibility had a different percentage chance.

 

[wlmartin]

I realized I probably need to take this back to formula since after doing some results, i don't think i put enough weight into certain alignments
I will come back to it but i tested the table method and it worked a lot simpler

 

[wlmartin]

@ delericho, as usual you are a giant amongst men!

So, at the end of this, you are going to end up with a list of alignments. As much as I'm a fan of the alignment system, this strikes me as a misuse of it, because the alignment system is only 9 buckets and what's interesting isn't the buckets themselves but what you choose to put in them. You are acting as if the buckets themselves are meaningful, when really they are just labels on a bucket and are defined by what's in them.

Rather than running the simulation as described, I suggest you back up and run the simulation with some initial abstract concepts. That way, the interbreeding of the abstract concepts will be suggestive of some new concept. Once you have the list of deities and their associated concept, you can place them in the bucket you feel is most appropriate.

I have used a system similar to the one you describe before, but not for the purpose of world building. Instead, I used it to determine the alignment of the children of PCs. However, the system was designed to also produce 'mutants' - whose alignment didn't match that of either parent - in a small percentage of cases. The list I used was something like: alignment of father, alignment of mother, one step more lawful than father (or lawful), one step more lawful than mother (or lawful), one step more good than father (or good), etc., ....lawful good, lawful neutral, etc. Each possibility had a different percentage chance.

So you are saying that the focus should be on creating mutations rather than similarities

So a LN and LE pairing could produce lots of LN, LE a in my scenario
In yours, its more likely I would want to produce chaotic and good alignment god children to sort of mess things up, create conflict and thus the timultius nature of the gods arises?

 

[Celebrim]

So you are saying that the focus should be on creating mutations rather than similarities

Well, not necessarily. However a world that was so neat and tidy that it never produced mutations would be setting an ideological expectation that the world is biased toward orderliness. What I'm saying is that in the real world, sometimes children aren't like their parents. Whether this is true of breeding gods, I couldn't say and its really up to the world builder to decide.

In yours, its more likely I would want to produce chaotic and good alignment god children to sort of mess things up, create conflict and thus the timultius nature of the gods arises?

I don't know what you want. Do you want a pantheon that feature more tumult and pettiness, or do you want a pantheon that is an orderly bureaucracy with each god fitting into a proper place and role?

More to the point, what do you think you'll accomplish by knowing the alignments of the gods? You're going to end up with something like this.

1st Generation: Neutral
2nd Generation: Neutral Good, Lawful Neutral, Neutral Evil, Chaotic Neutral
3rd Generation: Neutral Good, Lawful Good, Chaotic Evil, Chaotic Good, Lawful Evil, Chaotic Evil
4th Generation: Chaotic Good, Lawful Neutral, Chaotic Evil, Chaotic Neutral, Neutral, Neutral, Lawful Good, Neutral, Chaotic Neutral

After going through all that work, what are you really going to know? Who are these persons?