Free choice versus the laws of statistics – how does it work out?

Hello Rabbi,
In many places you talk about the existence of free choice even when human behavior can be described statistically. I would like to better understand your point:

1. If each individual's actions converge to a fixed distribution (e.g. 60% choose A and 40% B), what makes the choice "free"? The very existence of a stable statistical law seems to hide behind it a deterministic mechanism that ensures that the distribution is maintained.
2. At the macro level: The law of large numbers guarantees that as the sample size increases, the average outcome will approach the expected value. So how is it possible that the totality of a large group's "choices" could truly deviate from the pre-written probability? Isn't this a practical denial of the group's freedom?

In short: How do you extract an authentic margin for free choice, both individually and collectively, when there is a statistical distribution that predicts the outcome?
thanks!