Yeah, you're right. Besides, I was fixated on the 1 in 64mil odds, but those odds only apply if we declare a 6-roll sequence from the outset. But that's not how we think about patterns as humans. We don't think, "I'm going to roll my next six d20s and see if I land a 20 on all of them" Instead, we only start counting when we land that first 20, which effectively raises the odds 20-fold by eliminating anything but 20's from the first roll (I want to say the new odds are 1 in 3.2mil, but the math isn't quite so simple) So granted, the total sequence had a 1 in 64mil chance of happening. But from our perspective, it really was more like 1 in 3.2mil for the perceived pattern to emerge.