“In long-term political strategic forecasting, it’s been shown that experts are just not better than a dice-throwing monkey.” — Daniel Kahneman
After predicting the election results in 49 of the 50 states correctly back in 2008, Nate Silver became the go to expert. A position that was cemented in 2012 when he correctly predicted the results in 50 out 50 states.
However, after this election he’s had to answer the question “Why did Donald Trump win when you only gave him a 1/3 chance?” more times than I care to count.
His defense has been some version of the fact, and it is an objective fact, that a 1/3 chance is not insignificant. It is very significant, and you don’t even have to compare it to those that only gave him a 2% chance.
Let’s say you apply for the position of CEO at Apple. You have no qualifications whatsoever, but you do it for shits and giggles. By some miracle you become one of the two finalists, let’s call the other one “Jeff”.
Going into the office the experts give you a 1/3 shot of becoming the CEO.
But instead of having the board vote they tell you that the whole thing might as well be decided by rolling a dice. If it’s a 1 or a 6 you get the position, if it’s anything else it goes to Jeff.
That changes things doesn’t it?
We wouldn’t be surprised because we view a 1/3 chance with a dice as completely plausible, whereas a 1/3 chance for the presidency as “unlikely”.
It’s the same with the weather report. If they say there’s a 1/3 chance of rain, we go out thinking that it means we don’t need to bring along the umbrella.
There are two things we should start understanding:
- Polls are only as good as the data. In one of the interviews I saw Nate said that something like 10% of people (often less than that) asked actually respond. Are those 10% a representative sample of the population? Most likely not. It’s kind of like asking people coming out of the bathroom: “Did you wash your hands?”. How many will answer? And how likely is it that someone who didn’t wash their hands will actually tell you the truth?
- We should start looking at and understanding how forecasting and prediction actually works. If there’s only a 5% chance of rain in your area, that doesn’t mean it won’t rain. What it means is that given the same circumstances it WILL rain 5 out of 100 times. Today could be one of those 5 times.
Another aspect of why the odds varied was that when analyzing data there are different approaches. For example, Nate’s data favored recent polling results while Huffington Post’s favored historical polling results.
This becomes even more bothersome when we start talking about “averages”. Here’s a tip, whenever you hear a news report or someone presents you with a statistic or “fact” claiming that “the average…”, make sure to ask the question: “By ‘average’ do you mean the mode, median, or mean?”
Now, the problem we’re facing isn’t that the individuals and polls were wrong.
The problem is that we’re very likely going to cling on to whoever seems to hold the answers. We’re already looking for the next guru.
Take a look at this interview with George Packer on “The Daily Show with Trevor Noah”:
Within 10 seconds from shaking hands with him Trevor says: “You predicted the future.”
Thankfully George Packer tries to downplay it. But when he explains what he came to understand during the process of writing his 2013 book “The Unwinding” (it’s a great read!) it does seem like he knows something none of the polls took into account.
And that’s how we create these experts.
Now, I’m not saying that you’re going to see George Packer on every show talking about this, nor am I saying that he’ll be invited back ahead of the 2020 elections to talk about the growing sentiments in the U.S.
What I am saying is that we need to zoom out, understand that people have different areas of expertise and that none of them have all the answers. They’re experts in a specific area, and they’ve probably worked long and hard at attaining that.
While this doesn’t mean they’re immune to mistakes or misinformation we should still listen to them, and supplementing what they have to say with other sources as well.
So, why do we keep looking to these experts again and again?
Because we want certainty.
We can’t stand the fact that things aren’t so easily set in stone. We’re looking for guarantees in a world where there is none, and the answers only become “obvious” after the fact.
Whoever got it right, that’s who we follow.
To sum it up, here’s what you should take away from this:
- Be critical. The quality of the information is only as good as the source from whence it came. You should also supplement it by looking at the other side, actively search for information that contradicts it. Try to see if there’s consensus around the topic.
- Have an open mind. Not everything is black or white, there’s plenty of grey. There can be many different benefits and/or risks. Explore them and be willing to change your mind and adjust when things you thought were one way turn out to be another.
- Talk to people, and listen to people. Communication goes both ways, be willing to both challenge people (be critical) while also listening (have an open mind) when they challenge you.