I just remembered an old 'statistic' I once read about.

It seem that at one time, there was for several years and almost perfect correlation between the sales of alcohol and the salaries of college professors.

I believe it was used in a statistics course to demonstrate the dangers of drawing too many inferences.....