# Too Good to be True

Today is a topical post! I don’t do these very often, but something has been floating around the news a bit today and I want to make a larger point about it.

To keep this relevant for future readers, let me give you the quick background. In the 2020 Democratic Primary race, candidate Mike Bloomberg spent a lot of money on his campaign for the party nomination. Like… a lot of money. Like half a billion dollars. Then he dropped out of the race after mediocre performance, so take that as you will.

Then someone tweeted this gem:

So, okay. They’re bad at math. But honestly, I don’t care about that. Lots of people are bad at math. Being bad at math is actually okay if you’re good at being reasonable.

Consider the following situation: you’re attempting to calculate the weight of your cat. It’s often hard to get pets to stay still on a scale, so a common method is to weigh yourself, then pick up your cat and weigh yourself again, and then subtract. Usually you get a close enough figure for what you need.

Now, let’s say you did that, and when you did the back-of-the-napkin math, you accidentally made a simple error and instead of your cat weighing 11 pounds you calculate it as weighing 110 pounds. Even if you didn’t spot the error in your math, what you should immediately catch is the absurdity of the result. No reasonable person would accept that their cat weighed over a hundred pounds without raising an eyebrow, which should in turn tell you that your math must have been wrong, even if you didn’t think it was. You’d go back and check again because the result was so bananas.

Now, when someone doesn’t get the right answer to a particular problem they’re working on, there’s usually one of three causes:

1. They didn’t have the correct inputs/knowledge in the first place. It happens.
2. They made a mistake during the process and didn’t pay enough attention to spot it.
3. They were highly motivated to get the wrong answer.

That last one happens a lot. Sometimes the wrong answer is exactly the answer we want. “Can I afford to quit my job and retire early,” you ask yourself. You do some math, and you make a mistake. As a result of the mistake, you believe that your current retirement savings will be enough to live comfortably, when in reality you’ll be stretched very thin and you should work for another 5 years. But because you really, really wanted the answer to be “yes I can,” you don’t spot the mistake.

Train yourself to be suspicious of good news. Extra suspicious, in fact. I love dispelling folksy truisms, so let me reverse a classic one: You should absolutely look a gift horse in the mouth.

First off, it might be filled with Trojan soldiers, so there’s that.

But second, if you really want something to be true, you will be less cautious than your baseline. So you should get into the habit of being MORE cautious than your baseline when you get good news, so hopefully the two effects will roughly balance out and you’ll be as smart as you always are.

In the above tweet, the person should have known that one person being able to casually send a million bucks to every American is as patently absurd as your cat that you just picked up and stepped on the scale with weighing a hundred pounds. But it supported the worldview she wanted to have, so she wanted it to be true, so she didn’t check hard. If someone with the opposite worldview had done the same math and gotten the same erroneous result, they’d have scowled and done the math again – to them, the answer would have been too bad to be true.

If you can train yourself to be as skeptical of news you want as you are of news you don’t, you’ll make far fewer mistakes.