“1 in 64 isn’t a miracle, it’s clustering.”
Misuse of the science. It can only be clustering if every flip was with the same coin in sequence as you suggested the 1K coin test.
No, it’s still clustering. Clustering is random results bucking the statistical expectations. You expect somebody predicting coin tosses (including of multiple coins) to get the right answer 50% of the time. By getting it right 6 times in a row they bucked statistical expectations, but small sample sizes have a tend to buck statistical expectations, and some people read too much into it:
https://en.wikipedia.org/wiki/Clustering_illusion