About Being Bayesian
The amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it.
So goes Brandolini's Law, or the Bullshit Asymmetry Principle. That's the reason why memes, propaganda, jumlas and misinformation spread so easily. People say a lot of outrageous, often incrorrect things. And they're rarely held accountable. Philip Tetlock wrote that the predictions (and opinions) of experts have never been measured for accuracy. If they were, we'd find that the average expert is about as accurate as a dart-throwing chimpanzee.
Hans Rosling, too, in the introduction to his book Factfulness, points out that most people are often more ignorant about the world than chimps. A group of chimps randomly picking answers from a set of options would be right at least on average. Humans, on the other hand, are devastatingly wrong - all because of clickbait and misinformation.
Chimps might succeed on average because they have no memory of past failures and no hope for future success. But if a chimp can be a Bayesian, if it can replace its imagined certainties with degrees of belief, and if it can keep changing those degrees as more information comes to light... Let's just say we'll end up with a planet of different apes.
This newsletter is an attempt to be at least as accurate about the world as a chimpanzee.
Now, I'm under no delusions. I know that all the Tetlocks, Roslings and Kahnemans of this world - even if they had an army of Bayesian chimps - will fail against BS. Brandolini's Law is inviolable.
But there's certainly something liberating about accepting that. Accepting it means that there's an almost endless supply of BS to analyze and write about. It means we're freed from the burden of having to refute it all, and we can instead have fun with statistics and data visualization. We must imagine Sisyphus happy, and we may yet learn to laugh at the absurd.
That's what this newsletter is about. It's a monthly dose of the choicest BS, attacked with statistical reasoning, data stories and visualizations.
About Me
I'm a data scientist based in New Delhi, India. I've been telling stories with data for better part of the last decade.
My claim to fame is that on a few occasions, I've managed to change the opinions of an entire boardroom with with a single chart, after the spreadsheets had failed.
I get triggered when I think people don't know what they're talking about. I've been able to refute them successfully, too. But ultimately, being right about something is asymptotic. It may or may not come. It's a lot more fun to be first wrong, and then less wrong.
Why Subscribe?
You should subscribe…
primarily because, in the spirit of being Bayesian, I welcome corrections. The person in the XKCD sketch above is both you and me.
Moreover, you don’t need to be a statistician to be able to enjoy this newsletter. As long as you can compare two numbers in your head, I promise you’ll have fun.
Ultimately, the analytical treatment of stuff in this newsletter is just that - a treatment. It’s a means to an end. The real fun is in outraging against BS. If you’re on the internet, you know how entertaining outrage can be.
So, if you’ve read so far, please
