This seemed fuzzy to me. Thankfully, I work in a research institute, and there is a world-class mathematician sitting one office away who was willing to take a look at it.

So, if you live in this flood plain for fifty years, what you actually have are fifty chances of a 1% event. That's not the same as a 50% chance.

Imagine you are flipping a coin, and it comes up heads 99 times in a row. What's the probability it will come up heads again? It seems low, but it's actually still 50%, because the final flip (experiment) is completely unaffected by the prior 99.

To figure out the actual probability of a flood over a fifty year period, you'd have to know whether or not there was any correlation between the chance of a flood and the length of time between floods. I don't think there is, but unfortunately we don't have meteorologists here. If there is no correlation, nothing that increases the risk if, say, there hasn't been a flood in a decade, then your risk remains 1%, no matter how many years you live there.