That isn't how statistics work. If 10% of people are "addicts", and 20% of CEOs are "addicts", it's fair to say that there is correlation - despite the fact that there are still five times as many non-addict CEOs.
edit: Think this is actually the point you're trying to make, though I think you didn't quite nail it.
I mis-spoke when I wrote "expect 'addictives' to be CEOs just as often as non-addictives". I meant to write "expect to find 'addictives' amongst the population of CEOs at the same rate at which they occur in the general popluation" or something.
Would've been easier to write it out with math. :P
edit: Think this is actually the point you're trying to make, though I think you didn't quite nail it.