I love it how most people think the future can be predicted in an accurate and consistent manner, from Average Joes who consider themselves experts after reading a few articles on the Interwebs to PhDs who think their models are the best thing since sliced bread.

“But… but Joe Psychic correctly predicted X, Y and Z”, you might be tempted to say.

This is a common situation I come across when talking about how ridiculous it is to claim you can accurately/consistently predict the future, a situation in which people tell me I’m wrong by giving me an example of someone who got it right a few or even several times.

In my opinion, this way of thinking is flawed.

Don’t worry, I won’t “prove” this by turning this post into a PhD thesis, I’m simply going to share a simple example with you today.

1) Let’s assume the two of us want to flip a coin once.

I say “heads”, you say “tails”… we flip the coin and it’s “heads”, so I was right.

Question: did I get it right because I’m a brilliant forecaster or because I was lucky?

2) Let’s assume a coin is flipped not once but two times and that Joe, Alex, Tina and George are asked to “predict” the correct sequence.

Joe says “heads and then heads.”

Alex says “tails and then tails.”

Tina says “heads and then tails.”

George says “tails and then heads.”

There are no other possibilities.

The coin is then flipped two times and the outcome is “heads and then tails,” so Tina ends up being right.

Question: did she get it right because she’s a brilliant forecaster or because she got lucky?

3) If we flip a coin three times, there are 8 (2 to the power of 3) possibilities.

If we therefore ask 8 people to guess the exact sequence (heads-heads-heads, heads-heads-tails and so on) and don’t allow two people to have the same guess, one of them will inevitably get it right.

Question: did he get it right because he’s a brilliant forecaster or because he got lucky?

…

and so on.

The same principle is valid no matter how many coins we flip.

We can for example flip a coin 30 times and ask people to guess the exact sequence.

Guessing a short sequence such as “heads-heads” or “heads-tails-tails” might not be impressive but guessing a 30-flip sequence such as “heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-heads-tails” definitely is 🙂

But again, the principle I’ve referred to thus far is still valid.

With 30 coin flips, we can do the exact same thing as before.

There are 1,073,741,824 possible outcomes (2 to the power of 30), so we can ask a little over one BILLION people to “predict” the exact sequence and not allow duplicate predictions (so if the sequence you picked was already chosen by someone before you, we’ll force you to make a different prediction).

The result?

Out of those over one billion people, ONE of them is guaranteed to get it right and guess the exact 30-flip sequence.

Question: as impressive as guessing such an exact sequence may seem, did he get it right because he’s a brilliant forecaster or because he got lucky?

I’ll let you guys draw the conclusions, my job for today is done 🙂