how do you know if a stupid dice trick is actually interesting? is this something you can figure out ahead of time?
Unfortunately, the answer is mostly "not really". It's easy to think up Stupid Dice Tricks that are interesting in principle, but whether they'll be interesting in practice depends on whether you've correctly assessed how likely various outcomes are to come up at the table – or, put more simply, it depends on math.
In some cases, you can end up tripping over simple ignorance of how probability works; a classic example is game designers who are operating under the mistaken impression that all possible results of a sum of 2dX or 3dX are equally likely. I'm sure many reading this have seen indie RPGs with sum-of-2dX lookup tables for your character's background that produce a puzzlingly high frequency of professions that just happen to start with letters around the middle of the alphabet!
In other cases, the issue may be that you don't have a good grasp of how likely specific outcomes are. It doesn't matter how cool your dice trick is if the interesting results will only come up on combinations of dice that are so vanishingly unlikely, a typical group will probably never see them at all. As an illustrative example, imagine an RPG with conflict resolution based on poker hands, where you only score a critical success when you draw a royal flush (which, for the non-gamblers in the audience, happens one time in 649 739 in standard five-card draw); do you understand the odds your Stupid Dice Trick produces well enough to be sure you're not in this scenario?
(While the royal flush example is admittedly an absurd hypothetical, lesser versions have occurred in reality even in very high-profile games; fans of Exalted 3rd Edition could tell you all about it.)
Finally, the problem may be deviously subtle. Consider, for example, the dice-rolling mechanics in classic World of Darkness games. In brief, each challenge has a target number reflecting its difficulty, and the acting character rolls a number of dice equal to their skill rating; each individual die that comes up equal to or greater than the target number counts as a hit, with the total number of hits determining the degree of success. Dice showing 1s cancel out hits, and if there are still 1s left over after all hits have been cancelled, it's a critical failure.
This is fairly simple to describe, and not terribly challenging to carry out. The devil in the details, however, is that in certain very specific but surprisingly common situations, having a higher skill rating actually makes rolling critical failures more likely rather than less. Would you have guessed that this is the case based on the preceding description? If your answer is "no", you're not alone; it took years for anyone to correctly identify the problem and pin down exactly what was going on, and this is in a popular game line which had hundreds of thousands of active players at the time.
(Later revisions would mitigate the issue by changing the criteria for critical failure so that it occurs when a roll has zero hits with 1s showing before performing cancellations, rather than after.)
This isn't to say that any of this represents an intractable problem. There are much better tools available for figuring this stuff out in 2025 than there were in 1991; as long as you know where to find those tools, and have just enough knowledge to know what questions to ask them, these waters can be navigated with relative ease. You do, however, have to reality-check your ideas by running the numbers; unless you're a career mathematician with a very particular focus of study, you almost certain won't have an intuition for it. I have thirty years of self-taught combinatoric analysis under my belt, and when I try to just guess what might produce an interesting result space, I still fuck it up all the time.
