Political analysts use polls, betting odds, historic trends and other inputs to make their (usually wrong) political predictions. We all remember that “experts” said Hillary Clinton would win the presidency in 2016 (they gave her a 92% chance on the morning of the election), and that the UK would vote to “remain” in the EU (they gave that a 70% chance of the day of the Brexit referendum).

Both forecasts were wildly wrong.

There are many other examples we could provide. We use predictive analytic models also, but they’re much better than the norm.

We got the Trump and Brexit predictions correct. One of the factors we add that most analysts do not is anecdotal evidence. This could be anything from casual conversations with everyday people to counting lawn signs or observing bumper stickers.

Most analysts reject this kind of information because it’s difficult to quantify and put into equations. My view is the opposite.

Anecdotal evidence is the most valuable information you can gather because it comes straight from the people who actually vote and determine outcomes. Why use intermediary techniques when you can hear or see evidence directly? Another name for this is common sense.

A good example of a reporter who uses common sense is Sharyl Attkisson, the author of this article. She writes that Trump will be re-elected president next November. Her basis for this includes the extreme vetting Trump has already been through; (you may like him or dislike him, but there’s not much about him we don’t already know), and Trump’s track record of accomplishment.

Attkisson lists other factors that I would include under the heading of “common sense.” Professional analysts don’t think much of Attkisson’s method, but in my view, she has this exactly right. Reading this article is a good way to get away from the conventional noise and get straight to what’s happening in the real world.