Treasure tables part 3

02 Oct, 2024

Still trying to make sense of the D&D treasure tables but I think we're getting closer now. Someone on Mastodon pointed out that the weird spike at 1HD is an outlier because type A is specifically a "wilderness treasure" for big town-sized camps that's not meant to be in the dungeon.

I'm really struggling to find anything in the texts themselves that approaches explaining what each type is and what it's for. Possibly that's my own fault, to do with the nature of how I'm going about this - i.e. quickly, without reading things thoroughly, looking in the obvious places and not spending hours poring over the text for a clue. But also you'd expect that information to be in the obvious places so... whatever.

Looking at Holmes again Type A does appear to be for the Man-Types and it's a massive hoard. Similarly G is huge, and R is similar but missing magic items. They pop up at 1HD as well and appear to be for Dwarves and Elves, presumably in similar situations - i.e. this is the wealth of a population centre.

So let's strip them out, recalculate the values for each Hit Dice without those three massive types skewing things, and see where we land. I've once again put the XP thresholds for a group of 4 for the first few levels alongside them for reference.

That now makes a lot more sense, which is great!

If we make some of the same assumptions I made in the earlier posts - i.e. characters level at the same time, the party is always comprised of an equal number of characters who don't have hirelings etc., we always adventure on a dungeon level that matches our level and only meet creatures whose hit dice match the level they're on - then we can start to do some maths here. ¹

At 1st level we're looking at needing 1.5 average hoards to advance to 2nd level. Let's decide right now that we're going to round these up, because you can't find half a hoard but we're also dealing with average values. So we need two hoards to advance.

That's the same for going from 2nd level to 3rd level. But to go from 3rd to 4th level we suddenly need 8 hoards. And now we can get back to pacing. The first two levels of experience are about finding your feet, figuring out the game and your characters and the world that you inhabit. You get through the part of the game where it's more likely that you'll die a horrible and arbitrary death, and then you hit the meat.

You'll spend twice as long at 3rd level as you spent at 1st and 2nd levels combined. And that's really interesting! There's a similar amount of time spent at 4th level, and then we introduce dragons and the time spent to go from 5th to 6th is cut in half.

So how can I apply this to A Dungeon Game? Or how can it inform what I do with my tables, at least?

There are some different assumptions happening in ADG than in classic D&D. The first thing is that it's actually impossible for a brand new 1st level character to die immediately. You start play with a Scar, and when you hit 0 health you roll 2d6. If your roll is equal to or lower than the number of Scars you have then it's lights out, game over, roll a new character. You can't roll a 1 on 2d6, so you're always going to survive. Survival means you take a new Scar, and it's at that point that the potential for permadeath opens up.

So in terms of treasure and experience, even if I was concerned with getting players through that first level quickly (which I'm not sure I am), I wouldn't actually need to do anything to make first level pass faster than later levels. And since advancement and getting better at things isn't actually tied to level in the same way as D&D, the pacing aspect of levelling up is less important. (As I'm typing this I'm asking myself why, with all this being the case, I was so concerned with figuring out these D&D treasure tables, and I don't really have an answer other than that I always like to have some sort of framework to work in for this sort of thing).

The advancement that's linked to levelling in ADG provides a few benefits. You can cast more magic without having to exert yourself, you gain more hit points, and you recover from exertion better. Recovering from exertion is the big one here, because the better you get at that the more you're going to increase your attributes and the more characters increase their attributes the more trivial challenges become. So that's something that's worth bearing in mind, because if I'm not careful then at some point the game will simply break. ²

So I think my initial solution here is to err on the side of Not Enough Treasure on the tables, and keep level advancement "by the book" slower. GMs can always put more into their games (and I'd go so far as to say that they will always tend to, as well, because finding treasure is fun). Looking at the Holmes tables again, we've established that "by the book" advancement for the first 5 levels when represented as "number of hoards needed" goes something like: 2 > 2 > 8 > 8 > 4.

We could make ADG exponential, where we need 2 > 4 > 8 > 16 > 32, but that's getting silly.

There are obviously tons of options based purely in maths. We could use triangular numbers, in which case the sequence would be 1, 3, 6, 10, 15. The Fibonacci sequence would give us 1, 2, 3, 5, 8. Quadratic numbers would give us 1, 2, 4, 7, 11.

Much like the first example with the exponential increase, you always reach a point in a constantly increasing sequence where it just becomes ridiculous. There's no purpose in having an XP table that goes up to 10th level if it's functionally impossible to ever reach those levels. So there has to be a cap somewhere, a point where we say "you never need to find more than X hoards of an average value at the appropriate level to increase". 8 feels like a good cap for that, and that's also where the first 5 values of the Fibonacci sequence end. So maybe that can be some sort of basis.

Let's look at the XP thresholds from D&D all the way up to 10th level just to get that basis of comparison. As a reminder, these are the XP thresholds from Holmes to 3rd level and then from Moldvay/Cook for the rest (which is irrelevant because the thresholds for the first 3 levels are the same but in this house we stan Dr Eric Holmes). I'm assuming a party of 4 made up of a Fighter, Magic-User, Cleric, and Thief who for some reason all advance at exactly the same time, so I've taken the average of their individual XP thresholds for each level to give us an average Group XP Threshold.

Looking at this we can see that to get beyond 9th level becomes really fucking hard, and that's fine.

And honestly? After all the talk about different sequences etc. that we could use, I might just take this one almost whole cloth. There's a big part of me that wants the jump from level 2 to 3 to need twice as many hoards as from 1 to 2, so we'll make that 4, but otherwise? This is fine. And what this means is that I can reproduce this table with the numbers from A Dungeon Game to tell me how much XP value in silver a hoard should give out for each hit dice of monsters.

And just like that I have a framework to work from. The job now is to figure out how to distribute these values over a set of treasure types, then assign those treasure types to individual bestiary entries. That is, in fact, the actual work. It's what all this has been in aid of. But I feel much more prepared to handle it now, which is good. This has been a lot of fussing about with numbers and I think largely it's all been sort of meaningless, but it's something I needed to do to get to this point. Even if this distribution and these values are ultimately arbitrary in a lot of ways, in feels meaningful in that I've at least put some level of thought into it.

I don't have anything here to account for the things we stripped out - i.e. big hoards being kept by large population centres - but I'm fine with that for the time being. I don't have overworld/wilderness exploration stuff in A Dungeon Game at this point anyway (the answer to "Why not?" is in the title of the game), and that's an area that I can expand into in future.

Now to actually write some tables, I guess.

#adungeongame