I admit, for the most part, I really do believe President Joe Biden is very likely to be reelected this November. I agree that, so far, Biden’s team has not been the best at messaging, allowing far too much room for GOP nonsense to fill the void.

But then I remind myself this is just May. I recall a lesson from when I went to a candidate training program before my 2008 congressional campaign really kicked off. My instructor noted “normal” people don’t care about politics this far out from an election. Only weird people like, well, those of us in that room, care about politics this far out.

Stay up to speed: Sign up for daily opinion in your inbox Monday-Friday

That is why most political campaign experts will tell you to hold back from major messaging efforts until late summer or so. You can campaign early if you like, and send email and such, but voters won’t really start thinking about campaigns until much later political junkies do. You should spend these early months fundraising like crazy (I spent up to eight hours per day on the phone, which is truly soul-sucking work), signing up volunteers, and refining your plan for the fall.

Normal people just don’t care about this stuff this soon.

That being said, I’m still getting a tad nervous about some recent polling, and an interesting op-ed from the New York Times. The doom-and-gloom title of said op-ed is “Believe it Democrats. Biden Could Lose.” Uh-oh. And also in the same newspaper is a story that states former President Donald Trump currently leads Biden in the top-5 key states.

I’ve often compared the current political theatrics to what we saw back in about 1850 or so. The nation was profoundly divided and there was very much a national mindset among the political class that if you disagreed with someone, not only were you wrong, but you were also not a true American. Though I don’t think we are on the brink of an actual civil war again, the dystopia that Trump wants everyone to think they live is casting a surprisingly effective shadow across the American electorate.

Or maybe not.

Let’s talk about polls for a bit, OK? Do you all remember President Michael Dukakis?

No?

Well, back when 1988, then Gov. Michael Dukakis ran against the relative incumbent, then-Vice President George H. W. Bush and about this far out from the election, Dukakis was way in the lead. Bush Senior had to deal with an economic slowdown and other challenges that made his election doubtful at the very least. Heck, after the Democratic National Convention, Dukakis came out of the gates with a whopping 17-point lead, and yet he lost.

Now I get it is a tradition for whomever is behind in the polls to denounce said polls as wrong somehow. Well, that argument may be at least a little more accurate these days. Back when I was in grad school at the University of Michigan, I was fortunate to be studying at the top school in the nation regarding polls and polling. So, I ended up taking six courses in reasonably advanced statistics and polling, which gives me at least a little bit of confidence to say polls are increasingly getting things wrong.

Why? Well, back when I studied polling in the 1990s, you polled in two ways: you had some subset of your sample where you sent actual human poll takers to people’s homes (people don’t lie to poll takers in their homes as easily as they do on the phone) and the rest were carefully selected people reached out by telephone. Some fancy math lets you weigh the results to most properly reflect overall opinion.

I don’t have to tell any of you about the problem with landlines and polling these days. Few younger people have landlines (nor can they even dial a rotary phone, but that’s a different issue), and there are no phone books for cell phones.

The key to proper polling is to reach a sample that accurately reflects the opinions of the population you seek to understand. Cell phones make that much more difficult. Though polling organizations are hard at work finding ways deal with the cell phone problem, a representative sample is very tough to get. If you still want to call landlines, your sample will be skewed older, as older people like me are much more likely to have a landline than a kid. So, the polls themselves appear, at least to my eye, to be more likely to be off.

And though any final analysis of what the heck the Trump effect is will have to wait a generation or two to fully research, in my heart I have trouble believing Trump’s lies, his serial adultery, his business shenanigans and his multiple trials and likely convictions can fully pull the wool over American voter’s eyes a second time. Maybe, but I just can’t see it working, in that the opposition (to include lots and lots of independent groups that will buy ad time) will hopefully keep a steady stream of Trumpian BS front and center of the American electorate.

Oh, and abortion access may save the day for Biden as well.

Ultimately, the op-ed is correct, in that the Dems could hand the election over to Trump if they make the mistakes of 2016 and don’t take him seriously. His base continues to believe everything he says, and to be quite content with alternate facts, but his base is perhaps 25% to 30% of all voters, so he will clearly need to pick up many independents and others who are, at least currently, not that open to his various demagogical messages, and that may be what saves the nation in the end.

Stay tuned.

Hal Bidlack is a retired professor of political science and a retired Air Force lieutenant colonel who taught more than 17 years at the U.S. Air Force Academy in Colorado Springs.