While I do personally find tricks involving numbers only in a specific representation to be worth a bit less, often the underlying pattern of the trick generali

While I do personally find tricks involving numbers only in a specific representation to be worth a bit less, often the underlying pattern of the trick generalizes into a more interesting problem. For example, per another's link in these comments, this 'trick' works for 3 digits, but hits 1 of 3 possible loops for 5 digits. From this, interesting but likely useless questions can arise, such as finding an easy way to test for these loops, seeing if there is a way to calculate the loop without brute forcing it, and understanding the problem enough to know how much of this holds true when swapping to a new base. In general, most of this is just for fun and doesn't lead to anything serious. But sometimes a fun problem can be hard to solve, possibly leading to discovering something new, which ends up being applicable to more serious mathematics. Other times it can become a trap that just seems to waste time without ever leading anywhere, like the 3n+1 problem. I don't think this should be considered numerology, though I do think sometimes people treat tricks as if they have some more serious meaning that they don't deserve, at least not based on how they are presented. 3 Blue 1 Brown goes into the spiral pattern of the primes as something that appears to be deep, but ends up being an unique way to present an otherwise boring tidbit about prime numbers.