Disclaimer: The law I’m talking about here is the law of averages. As it is unenforced, I can celebrate its law-breakers with a clear conscience. Breaking those laws which are enforced in your municipality is badwrong fun, don’t do it kids!
In my last post, I was talking about intended consequences of the Adventurer Conqueror King design (increasing player agency by incentivizing the risk/benefit calculations that go into a decision to track a wealthy monster back to its lair, but sneak past a poor one) and how and why we’ve developed guidelines to support this (like that four experience points should be awarded for recovering treasure per XP earned in combat). That led to worrying about other cases, like the challenge rating system in 3E, in which similar guidelines produced undesirable consequences when gamers took what were meant to be suggestions and turned them into holy writ.
Justin Alexander, whose essay on fetishizing balance is one of the cornerstones of my thinking here, commented on my post to say (in part):
By offering equations which could be “solved” (whether for encounter-building, monster-building, or character-building), 3E offered the illusion that designing a good scenario was just a matter of “figuring out the math” and plugging in the right numbers. The encounter-building, monster-building, and character-building equations in 3E weren’t actually designed for that, of course, so the result is kinda sub-par. But 4E doubled down by specifically trying to take the DM “out of the equation”: If you put together the right mix of monster roles; solve the encounter-building equation correctly; and use the right number of encounters-per-day… then supposedly you’ll end up with a good adventure. ... [In] trying to provide such equations, 4E ends up unnecessarily constricting the game to a very narrow (and very bland) definition of “acceptable”.These kinds of constrictions are exactly what I wanted to avoid when I'd been saying that I see ACKS' guidelines as ways to make it easy to see what the normal outcome would be so that you can identify or design the exceptions that are unusual and thus exciting. Thinking about Justin's comment made me realize that although we share the interest in "figuring out the math" that characterizes the latest wave of mainstream RPG design, there is an important mechanical difference in the way we do it that emphasizes that lines are there to be colored outside of.
There are two stages to my work in setting up the monster/encounter/treasure type spreadsheets. The first is using equations to calculate how much treasure a monster of a given level should have. This is the same kind of figuring out the math that goes into 3E’s treasure per encounter level table, or 4E’s treasure parcels. (It is important to note that in ACKS, the treasure guidelines stay closely associated with monsters, so that the treasure type assigned to a creature says something specific about the way it gathers wealth in the game-world. 3E partially and 4E fully abstracted expected treasure away from individual monsters, so that they became a free-floating mechanic dissociated from anything in the world- but here we venture into the territory of another of Justin’s famous essays, and away from my current point.)
Then after having carefully calculated the “right” payoff from defeating a monster, my spreadsheets help me figure out what series of dice rolls will tend to produce this amount. Sometimes when I’m in the grip of a mania for precision, this seems crazy. If I know that defeating a hydra ought to yield 5,000 gp, why should I express that as a 30% chance of 1d4x1,000 ep plus a 25% chance of 1d6 x 1,000 gp plus a 25% chance of 1d4 brilliant gems plus a 50% chance of 1d6 pieces of jewelry? These dice rolls make it possible that the adventurers will get no reward for their victory, or one that is substantially greater than average. Wouldn’t it be better to just take the “correct” value and express it as a guideline, as with wealth per level or treasure parcels?
I didn’t have an answer to that question until now, which shows the wisdom of our general approach to avoiding unintended consequences: not changing things that worked in our source inspirations without a really good reason. What I’m realizing, though, is that the way the legacy treasure tables assign rewards by using the sum of a lot of individual probabilities invokes the law of averages to set up appropriate expectations. Most hydras in the world have some treasure in their lair, and adventurers can make plans based on this knowledge.
However, when the dice hit the table, quite often they are going to break the law of averages - and breaking the law is fun. The way treasure generation is set up, so that the expected result comes up most of the time but not always, automatically tends to generate another of the rhythms that are satisfying in old-school play: nothing special, nothing special, nothing special - whoa what’s up with this?!
Guidelines for treasure payout that don’t involve randomness require the GM to create interesting exceptions. That in turn encourages the players to think that any exception in either direction is the GM violating the game’s social contract, whether that’s by withholding their deserved reward-parcel for facing a challenge or by going soft on them and giving out so much Monte Haul that future challenges won’t be satisfying. (Note that this latter also results from new-school games making it easy for characters to spend their wealth to directly enhance their personal ability to overcome challenges, and discouraging spending on anything except magic items that enable bigger overcoming-challenge payoffs).
By contrast, a well-designed random treasure table spits out exceptional cases like a one-armed bandit: just often enough to keep it exciting. When a run of snake-eyes means I roll up a dragon with no hoard, the players know it’s not me that hates them, it’s the dice. Better still, all of our imaginations go to work on the problem: who might have stolen this dragon’s treasure? (Note that the canonical answer from the Hobbit is why I think dwarves have such a fantastic treasure type).
For the players this means how can we track down the thieves and jack them instead? (See above on murdering dwarves being the road to riches.) For me it means might the dragon seek to join forces to regain its hoard? Maybe I don’t need to think of this at all, because the players will suggest that possibility first. Even if it’s something I might have come up with on my own, I find it much more fun when breaking the law of averages means I have to think on my feet the same way players do. And building these exceptional cases into the system so that they arise in the moment of play means that they’re much easier to integrate into the flow of events. If I was using a treasure parcel system, it’d take pre-planning to come up with the scenario where the dwarves have robbed the dragon, with all the attendant risk that something will happen during the session that invalidates my clever plan. Better to have the pre-planning happen on the level of designing a table that will encourage this kind of improvisation - which is to say that this is one of the ways ACKS is striving to answer the question Gygax and Arneson raised when they thought the original Dungeons & Dragons had said enough: “Why let us do any more of your imagining for you?”