I've noticed a trend between reds and blues. Reds favor trying to model women and relationships, finding patterns to use in their lives. Blues don't like this, saying that you can't expect any given person to act a certain way based on a few key demographics.

One interesting line I've seen from reds when I press the topic is that even if not entirely accurate, they are still useful. That you need to find an answer. Blues will admit there are gender differences freely, but we don't think they can be used that much. Averages are just averages, and with a high amount of variation they can't be applied to individuals that well. Most physicists are men, but if you were guess that a random physicists was a guy, you'd be wrong 20% of the time. A blue sees this and understands that there is a real possibility that our model of physicists being men will mislead us, and we tapper our reliance according. We realize that further input is needed before we can be too confident. We are willing to accept uncertainty and say "we don't know".

And yes, some of go too far in the other way. Some of us seem to act like there are no trends at all. But that fits I this too. They are willing to say "I don't know" very fast.

I looked up the number of physics degrees awarded for my example. It's a pretty solid number. But for most of these topics we don't have good numbers. Reds seem to happy to build their models on personal observations and anecdotes. And they seem rather committed to their models. They don't much explore the possibilities of them not fitting the model, even though there is a real chance any given woman won't fit. That is, reds don't like uncertainties and unknowns. They want to have an idea of who a woman before ever talking to her. They don't want to consider that she might not be who they expected. Blues, on the other hand, are much more willing to accept uncertainty and unknowns. You might call it special snowflakes, but it's just realizing that even the best laid models break down on small scales.