~50% divorce rate is dated information, it's much lower and decreasing [0]. There's also the fact that the divorce rate amongst the college-educated (which I assume includes the software developer in question) has always been much lower. Something else that throws off divorce statistics is serial divorcers, a significant cadre who have divorced two or more times.

[0] http://www.nytimes.com/2014/12/02/upshot/the-divorce-surge-i...

divorce rates among upper income, educated folks is low. . . closer to 10-20% range. . . and to be fair some matches are just plain bad or green card marriages.

That wasn't the point of my post.

My point was that, no matter how you fudge the numbers, it's impossible to chalk up bad divorces as outliers. Outliers are defined as 3 standard deviations from the mean in a normal distribution, so the parent was implying that bad divorces occur at a rate of 0.1% or less--which is off by at least one order of magnitude.

> Outliers are defined as 3 standard deviations from the mean in a normal distribution

I kind of agree with your overall point, but that definition is wildly incorrect.

Hey there, my green card marriage is coming up on 18 years strong..

that 50% thing is not really true any more. http://bestdivorcelawyer.co/family-law/divorce-rates-higheri...