> In my mind, what makes something a statistically safe assumption would mean that, more times than not, you'd be right.

I assume I should buckle my seatbelt.

Not every car ride results in an accident. But enough do.

It's safe to assume you won't get in an accident.

You _should_ buckle your seatbelt anyway because it's low effort, high reward in the unlikely case you get in an accident.