I'd guessed you'd done the same thing with supermoons and checking the chance of any 6-day window having at least one quake. The numbers for both should have multiple quakes in roughly the same proportion.