The current roll is entirely independent from the previous roll. Thus it doesn't matter what you previously rolled for the purpose of knowing what you're going to roll. That you have rolled 1 ten times in a row doesn't mean you're due for a roll that is not 1. Having rolled 1 ten times in a row, it is no more or less probable to roll 1 an eleventh time than it is to roll 7 or 15 or 19. Any series of 11 specific values in a specific order is exactly as probable as rolling eleven 1's in a row.
In order to test if a d20 is actually unbiased, we need to record thousands of rolls and check that the trend over time is for all results to come up equally. Testing whether a d20 is truly random... Now that's not fun at all, because what does "random" actually mean? But as it happens, we know that Larian's implementation is not actually truly random, it's merely pseudo-random to a degree that immitates random well enough.
