Why We’re Blind to Probability
The idea that something can be likely and not happen, or unlikely and still happen, is one of the world’s most important tricks.
But let me tell you about a common problem. I’m as guilty of it as anyone else.
It’s that most people understand probability, but few actually believe in it.
Most people get that certainties are rare, and the best you can do is make decisions where the odds are in your favor. They understand you can be smart and end up wrong, or dumb and end up right, because that’s how luck and risk work.
But almost no one actually uses probability in the real world, especially when judging others’ success.
Most of what people care about is, “Were you right or wrong?”
Probability is about nuance and gradation. But in the real world people pay attention to black and white.
If you said something will happen and it happens, you were right.
If you said it will happen and it doesn’t, you’re wrong.
That’s how people think, because it doesn’t take much effort to think it.
This happens during elections. Nate Silver can say there’s a 72% chance Hillary Clinton will win. Then when Donald Trump wins people say, “Nate Silver was wrong.”
It happens during pandemics. A headline might say, “Experts predict 200,000 deaths by August.” What it should say is, “Experts have 95% confidence that the number of deaths by August will be between 110,000 and 290,000,” or something like that. But if the number of deaths by August turns out to be, say, 120,000, people will declare the experts wrong. If the number turns out to be 350,000 they will be called disastrously wrong – even though the forecasters knew there was a one-in-twenty chance of that occurring.
It happens in investing. Predictions that come true bring invitations to be on CNBC. Predictions that don’t bring client redemptions. Black and white.
We can’t just say this happens because people are dumb or lazy. Probability gets ignored for three reasons.
1. People don’t want accuracy. They want certainty.
A lot of what goes on in the financial world is an attempt to rid yourself of the painful reality of not knowing what’s going to happen next. Someone who tells you there’s a 60% chance of a recession happening doesn’t do much to erase that pain. They might be adding to it. But someone who says, “There is going to be a recession this year,” offers something to grab onto with both hands that feels like taking control of your future.
When you realize that making people feel better is more appealing than giving people useful figures, you start to see why probability is rare.
2. It takes too long for a sufficient sample size to play out. So everyone is left guessing.
Let’s say you’re a 75-year-old economist. You started your career at age 25. So you have half a century of experience predicting what the economy will do next. You’re as seasoned as they come.
But how many recessions have there been in the last 50 years?
There have only been seven times in your career that you’ve been able to measure your skills.
If you want to really judge someone’s abilities you would compare dozens, hundreds, or thousands of attempts against reality. But a lot of fields don’t generate that many opportunities to measure. It’s no one’s fault; it’s just the reality of the real world is messier than an idealized spreadsheet.
It’s an important quirk, because if someone says “there’s an 80% chance of a recession,” the only way to tell if they’re right is to compare dozens or hundreds of times they made that exact call and see if it came true 80% of the time.
If you don’t have dozens or hundreds of attempts – sometimes you have one or two – there’s no way to know whether someone who says “75% chance of this,” or “32% chance of that” is right or not. So we’re all left guessing (or preferring those who profess certainty, which is easier to measure).
3. Distinguishing between unfortunate odds and recklessness is hard when risk has painful consequences. It’s easier to see black-and-white even when the odds are apparent.
I valeted at a hotel in college. We parked 10,000 cars a month. And we banged one of them up every month, like clockwork.
Management found this atrocious. Every few weeks we’d be scolded for our recklessness.
But one accident in 10,000 parks is pretty good. If you drive twice a day, it’ll take you 14 years to park 10,000 times. One bent fender every 14 years is a driving record your insurance company won’t bat an eye at.
But try explaining that to your boss who had to file so many damage reports they got to know the insurance claims adjusters by name. You’ll get no sympathy. It’s so much easier to say, “You guys are clearly being reckless. Slow down or you’re fired.”
The same thing happens during bear markets. You can show people that the market historically crashes every five to seven years. But every five to seven years people say, “This is wrong, this sucks, this feels broken, my advisor screwed up.” Knowing the high odds of something happening loses its meaning when that thing happening hurts. Probability goes out the window.
Same in an investment portfolio. If you make 50 investments you will probably do OK on half, terrible on a quarter, and great on a quarter. That’s what success looks like. This is true in various degrees whether it’s blue-chip public stocks or venture capital. But even if you know that the majority of investments you make won’t be great, anyone looking to question your skills – sometimes it’s you questioning your own – will have plenty of examples to criticize. Odds that are in your favor but still hurt are hard to distinguish from odds that aren’t in your favor.