So very presidential.

As you may have heard, the President of the United States has an active personal Twitter account. We all fondly remember the days when the tweets weren’t national policy, but rather the musings of the guy on the TV show with the inexplicable hair.

Friendly relationship advice for Robisten? Check!

Musings on the effectiveness of Diet Coke versus Coca Cola Classic? Check!

Solemn remembrance of the fallen of 9/11? Checkerino!

But now, with Mr. Trump the duly elected president of these United States, tweets are presidential pronouncements. Not surprisingly, his tweets — before, during, and after the 2016 campaign — are littered with logical fallacies. Let’s check out a of them.

AD HOMINEM

By far, the most common fallacy in Trump’s tweets is ad hominem. This is when you attack your opponent’s character rather than addressing the substance of your opponent’s arguments. This is literally the bulk of Trump’s campaign — remember hits like Lyin’ Ted, Little Marco, and Crooked Hillary? — and I could write tens of thousands of words on tweets containing the same flaw, but, since we don’t have all year, let’s just look at a few:

— There’s this gem, where Trump retweeted a photo meant to compare the looks of Lyin’ Ted’s wife unfavorably with his own.

— Or this one where he’s totally not calling Megyn Kelly a bimbo.

TREATING CORRELATION AS PROOF OF CAUSATION

Everyone knows that the mere fact that two things happen together does not necessarily mean that one caused the other. Well, everyone but the president:

— In this tweet, Trump assumes without supporting the idea that his election victory caused a jump in consumer confidence. Thanking himself in the third person is a nice touch.

INAPPROPRIATE APPEAL TO AUTHORITY

It’s fine to cite someone who’s authority in a field is well regarded when you’re making an argument. The big problem is when you stray from that well regarded authority standard.

— In this rich example, Trump cites “an extremely credible source” as proof that then-President Obama’s birth certificate was a “fraud.” A bald assertion like this doesn’t cut the mustard, but you should know that the quotes around “extremely credible source” are Trump’s, not mine. Oy vey!

SAMPLING ERROR

There is often overlap with this error and Treating Correlation as Proof of Causation because clinical trials meant to prove causation operate on sampling. The most common is an unrepresentative sample. There are also samples that are too small.

— This stinker has both of those problems. Here, Trump is mentioning some kid (assuming this kid exists) and concluding that vaccines cause autism. He goes on to talk about “many such cases,” but it’s quite possible that those cases are not representative of the wider population. One must use a randomized sample to confirm something like this, and that’s not what’s going on here.

QUESTIONABLE ASSUMPTION

If there’s a good reason to believe that one of the premises supporting your conclusion is false, there’s a huge hole in your argument.

— In this whopping whop-hole of a whopper, POTUS attacked millions of people marching the day after his inauguration wondering why they didn’t vote. Well, there’s no reason to believe these people didn’t vote. Even assuming high estimates of as many as 3 million people marching in the Women’s March, it’s entirely possible that every single one of them cast a vote in an election in which about 120 million people voted. Sad!

USING TOO MANY DAMN EXCLAMATION POINTS

Okay, this isn’t a fallacy, but seriously dude. Cut it the f*ck out.