I heard a market prognosticator on the radio at the start of June, claiming that the market was going to drop soon. Well, good for him; people make such predictions all the time, and they're right roughly half the time.
The interesting thing about this prediction was that this fellow made the argument that we're due for a big correction because "nothing goes up forever". I agree completely that the chance of having no down-days on the market approaches zero as your time horizon extends farther into the future, but that general argument, based on the law of large numbers, most certainly does not support his conclusion that the beginning of June was a good time to sell stocks.
It doesn't even matter that the stock market is down 3.8% from the day he made his call. He was still wrong.
If I toss a coin and get 10 heads in a row, I could make the statement "you can't get heads forever" and I would be entirely correct. If I went on to make the prediction that the next coin toss is now more likely than before to come up tails, I would be dead wrong, even if the next coin toss did in fact come up heads. You can't credit your pet theory for your own lucky guess.