Robert Bohl I meant something different than that. Most of RPG play state is unquantified, subjective, and evolves in a free-form way. All of this is highly idiosyncratic to a group’s pre-existing play skills, habits, beliefs, creative output. There’s an ongoing back and forth between this stuff and the game procedures, which (when the two combine) necessarily produces different outcomes at different tables.

This is a distinct phenomenon from rules clarity, self-consistency, whether the group understands the rules, or even whether they like the rules, their habits, or the outcomes of play. They’re stuffing different ingredients into the game’s rules, so different outputs come out.

I think that through extensive playtesting with a wide variety of groups, a designer can get a glimpse of the distribution of player habits, and can start to see what happens when game X is mixed with them. This might lead them to change the rules so that the outcomes are slightly more consistent for a wider range of habits (or more likely, just aimed at the average a little better).

But it’s still going to be a regular occurrence that a group correctly understands the rules, mixes them with their own creative habits and group dynamic, and doesn’t like the outcome. In this case, hacking the game seems a completely sensible thing to do.