If you are going to suggest someone has effectively said something meaningless, why not break it down properly and calculate the error bars?

I'll be just as lazy...4x the number of male duds across that number of observations? It's going to be statistically significant.