"Forecasts are nice because human beings crave certainty," Brown said. But forecasts are inherently flawed, he said, because things often don't go the same way they did in the past. And a forecast can be the enemy of an investor -- if you're always chasing the last expectation.
"You literally cannot assemble a data base of data and have it spit out the answer to what things are going to look like in a year, three years, five years. It's just not literally possible."
Remember all those forecasts that predicted gloom and doom for stocks if Donald Trump was elected president? The consensus was that Trump would be a disaster for the stock market.
Or what about the Brexit buzz in the summer of 2016? Again, the stock and bond markets were convulsing over the prospects of the United Kingdom leaving the European Union.
And don't forget that the experts said people in the U.S. wouldn't vote to leave the EU. And the polls indicated that Trump wouldn't be elected president.
But the people of Britain voted to hit the exit doors and a few months later @realDonaldTrump was even more of a must read on Twitter. Trump was elected.
The talking heads got the votes wrong and stocks went up, not down, in both cases.
"Polling is a whole separate category of nonsense," Brown said. "With polls, people are virtue signaling."
They might say what they expect the pollster wants to hear or what, maybe, they should say.
"Think about some of the stances President Trump took during the campaign," Brown said.