Does anybody remember the election in 2016 — because one of the same patterns is emerging. The polls are giving the Democrats an 80% probability of taking back the house. Uh oh, Trump may have them right where he wants them.

In 2016, polls overwhelmingly projected Hillary Clinton to win the election. Nate Silver’s forecaster, Fivethirtyeight.com, before Election Day gave Clinton a 70% probability of winning, and he has been a very accurate forecaster of elections. By 11 P.M. that night, after Pennsylvania was called for Trump, her chances had dropped to 5%, and that was being generous. To focus attention, it’s better to think that she started with a 1 in 3 chance of losing, which is what happened. The Democrats have a 1 in 5 chance of not taking back the House. Is that a cold shower or what?

A probability of 80% assumes that everything goes as the Democrats expect. If things do not go perfectly, the percentage diminishes rapidly. The polls get a whole lot more accurate once voters go to the voting booth. Forecasting polls are a snapshot and contingent; who answered the phone and who didn’t. They are inexact and can’t be treated as a foregone conclusion. Don’t look at the 80%; look at the 20%. That’s the key.

Don’t hold your breath until the House turns blue. Get out and canvass, make phone calls and get everyone to vote. If you’re in a safe district, assuming there is one, work for a district that is contested. Participatory democracy means pushing beyond complacency and escaping your own gravity. Step out of your front door and roll up your sleeves.