The world is a big place. With airplanes and the internet, we can now see all of it. This is something new in recent human history.
We aren’t hard-wired to process the huge amounts of information that describe our modern world. This quote sums it up pretty well:
A single death is a tragedy; a million deaths is a statistic.
To deal with this quantity of information, we need concepts from statistics and computer science. This article is the first in a series I’m planning on “statistical thinking”, where I translate those concepts back into plain English, for everyday use
Let’s start with the basics.
What is a probability?
It’s actually very simple. If I went to sleep after 3am on 12 nights in the past 30 days, then “the probability that Mason is awake after 3am” is 40%.
Here’s how you calculate that. 12 (number of nights I was up past 3) divided by 30 (total number of nights in the period) equals 0.4, which is the same as 40%.
Probability is the observed frequency of an event, in the relevant set of circumstances.
(There’s a group of people, called Bayesians, who will disagree with me on this definition. But that’s just because Bayesians don’t know how to keep their opinions to themselves—zing!)
If you drive from San Francisco to the South Bay on highway 101-S in the morning, the chance you’ll hit traffic is 100%. In recent history, it’s always been true. That doesn’t mean that it’s impossible for there to be no traffic tomorrow. That can absolutely happen. Probability is not a guarantee about the future. It’s a description of the past.
This is an important point. When someone tells you the probability of something happening in the future, they are 100% full of shit. Most likely, they’re trying to sell you something. If not, they’re trying to get you to change your course of action.
Decision making in uncertain situations is another topic. It’s related to probability. You can use probabilities as inputs into your decision process. But there’s a lot more to it. One reason why, is George Soros’s concept of reflexivity.
Stay tuned for more articles like this on statistical thinking. You can follow me on Twitter for updates when I post new articles.