Base rate first. Base. Rate. First.

Here is a tip: in making any prediction, look at the base rate first. This will significantly increase your odds of being right.

For example, when thinking, will this startup succeed? Don’t think, yeah because the founder is charismatic or no because the market is not really tested. Think this first: Only about 10% of startups ‘Succeed’, and if it’s consumer hardware startups, the number is around 3%. Once you know the base rate, then think about the other variables, given the low base chances, how much different this startup actually is, to defy the rate? To land among the winners?

Here is an example, from an error I made. As I mentioned here, I made a prediction about Haruki Murakami winning the Nobel prize. I was driven by these things:

  • Media speculation about Murakami being a favorite. News articles saying that he is the Leonardo Di Caprio of the Nobel prize for Literature. (This shows how bad news can be for your understanding of the world, but that’s a different story)
  • Odds of bookies. Some sites published odds of bookies and Murakami was among the top 3 favorites. I have absolutely no idea how reliable the bookies are, but I assumed they might know the internal workings of how the Nobel Prize for Literature committee works.
  • My personal liking for the works of Murakami. I am a huge fan and love many of his books, especially, ‘Kafka on the shore’, it’s magical. Why should this genius not get the prize? I am a fairly well-read person and like to think that I can discern good literature from bad.

But even then, I asserted the prediction odds of 20%. This looks a fairly conservative estimate, right? Take a moment to stop and think about it, is it conservative, is it safe? Is it even?

Now let’s look at the base rate: The Nobel Committee received and approved 195 candidates for the Nobel Prize. And they do not reveal the nominees. This makes the base probability 1/195, or 0.51%.

When I thought it would be around 20%, I was essentially asserting that Murakami was 20/0.5 = 40 times better than the other nominees. Nominees I did not even know of. Of course I was very, very off-the-mark even with a 20% chance. If I had thought about the base rate, my estimated odds would have been much lower. Now, if I had known about the other nominees, and also had read their works, and also was of a good enough literary taste to discern what was better, then I would have been in a better position to predict the odds. But alas, I was driven by my own liking, and of course, the news.

It is good that this was just a playful exercise. I had no money riding on it, but in bigger decisions it’s good to remember: base. rate. first.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s