It's too random a process to be precisely answerable about a specific data point or two.
One could argue that this is a doing-something as opposed to a saying-something, and thus more substantive. Or perhaps people want some good news to believe in? I don't know - one can make up lots of just-so stories about these things (see paragraph 1).