Agreeing.
In another life, I was a statistician and did a brief stint teaching statistical analysis and probability (as well as some sociology classes on the effects of the Media) at the University of California. If you want to talk about manipulation to make a point, I can talk to you about statistics all day. The reason I bring this up because there is a concept in statistics (this is a real thing by the way and is generally seen in multiple regression analysis) called "interaction between variables" and these interactions can lead to significant misunderstandings.
For example, IF you had a significant economic collapse and IF, by some wild chance, you had 3 or 4 significant dependent (predictive) variables and IF you used an independent ("predicted" for lack of a better word that I can think of right now) variable that made sense and IF you didn't leave anything significant out of the equation.......(okay, I think you're probably getting the point now.....this MAY not be completely reliable when it comes to determining the effects of variables in an economic collapse), the variables you selected could interact in very unexpected ways. That is to say, they might not have linear relationships with each other. Adding one variable or deleting one variable can significantly change the interactions between the other variables. This could lead you to dramatically over or understate the effects of the variables you used to predict some occurrence.
That and the fact that when people use after the fact variables that fit what has already occurred, it renders their predictive value virtually meaningless because there are a ton of things you can pull out of the air that will fit the model and some may have no value whatsoever.....think about "Every year that the Redskins win at home on the weekend before the presidential election, this or that party wins....every year they lose, the other party wins......" These are after the fact observations that likely have no effect on anything. Now statisticians (generally speaking) aren't quite THAT sloppy, but some do a pretty good job of obfuscation.
OK. Everyone can wake up now.....