So was parent commenter's experience just that non-representative or is something screwy going on with how the data gets recorded?

I think it's probably just very, very unlikely that any individual person individually knows a representative sample of divorced couples, and within what sample there is the original commenter noted that the father contesting custody was a subset.

I think people also maybe aren't primed to think of this in terms of "What percentage of people who contest custody receive custody, by gender?" They're just thinking to themselves "Of the couples I know who are divorced, in what percentage does the woman have custody?", and that is going to be very large because a very small percentage of men contest custody. Anecdotally, almost every divorced couple I know has had child custody with the woman (I am in Australia, for the record). The much smaller subset of "divorced couples where I know the father contested custody" are two father's with custody and one mother. Obviously, those are also the only divorced couples I know where the man has custody - no court is going to give custody for a child to someone who doesn't want it, ideally. I also know a lesbian couple who've gone through a custody dispute but I don't believe that really has any impact here.

Unless GP is a marriage counselor or family-court judge, I am wondering how many anecdotal child custody cases they would have witnessed.

From what I’ve read so far, there’s a lot of variation between states. The law used to explicitly prefer the mother, and that may still be the case in some states. Beyond that, n=2 isn’t a great sample, but there’s definitely going to be bias when it’s a friend sharing only their side of the story. Like how anytime someone you know gets into a car accident, it was definitely the other guy’s fault.

> Overwhelmingly, the fathers are giving up custody and not fighting for it.

Couldn't that be confirmation bias - if you don't think you are going to win, and it will cost lots of money - why fight?

Which is why when fathers fight for custody they may win more - because they have a stronger case.

> Couldn't that be confirmation bias - if you don't think you are going to win, and it will cost lots of money - why fight?

Confirmation bias is defined as "the tendency to interpret new evidence as confirmation of one's existing beliefs or theories."

What you're describing is not confirmation bias, because there is no "new evidence" which one could interpret as confirmation of their existing beliefs or theories. If the father doesn't even try to obtain custody then they haven't discovered anything new that would reinforce their existing beliefs.

It's not confirmation bias for the men, it's more for the researchers: Selecting data on who contests custody as a proxy for who wants custody. If there is a gender bias in the courts, that might not be accurate.