That's an explanatory bug on my part.

I have a little "mad-lib templatized" set of sentences for reporting on various stats, and that one was supposed to be sort of related to the law of small numbers. For example, shorter runs mean there is a larger probability of getting lopsided distributions (different from pure 11%).

It's supposed to demonstrate that even though you used a single number a disproportionate number of times, it's not necessarily statistically significant since shorter RNG runs can produce the same effects.