Friday, December 8, 2017

The Pareto Principle and the OSR

I've been spending a lot of time on wikipedia recently.  The other day I was reading about power laws, allometry, and the Pareto Principle, and it struck me as both a useful rule of thumb for the lazy simulationist and a convenient explanatory factor for some pieces of OSR weirdness.  Phrased simply, the Pareto Principle states that in general, 20% of causes account for / lead to 80% of effects.  Typical examples include aggregation of wealth / property (where in most countries, 20% of the population controls 80% of the land / wealth) and software bugs (where 80% of bugs spring from 20% of root causes, such that if you fix the most-common 20% of bugs, you also resolve roughly the next 60% incidentally).

One OSR peculiarity made believable by this observation is the lair / nonlair distinction.  Roughly 20% of encounters (lairs) have the vast majority of the treasure (certainly treasure-from-monsters) in any given dungeon level.  From the perspective of even, predictable progression this is a dubious choice, but from a simulation perspective, with monsters aggregating treasure or paying tribute off-screen, it seems solidly justifiable.

The Pareto Principle may also have applications in constructing believable causal / narrative structures from randomness.  You roll a result on a random table, and you're not sure why it's true.  Pareto suggests that there's an 80% chance it follows from one of the Big Causes of your world.  Roll a d10, on 3+ link it back to something already known, on 1-2 add a new minor causal agent unfamiliar to your players.  You could go even further with self-similarity on that 80% and balancing the sizes of your sets of big and small causes, but for a first cut that minimizes effort, tends to conserve campaign capital, and introduces occasional surprises, this is probably a reasonable solution.

If I were feeling extremely motivated, it might be interesting to look at Pareto as it applies to ACKS' economics / XP model.  But I am not; they're probably already reasonable consistent, given Autarch's economics background.

Other plausible correlates in gaming, absolutely unsupported by data but plausible to my eye - 20% of the bloggers get 80% of the pageviews, 20% of the systems get 80% of the playtime, 20% of the campaigns get 80% of the sessions, 20% of the gamers run 80% of the games.