Everybody Lies -> Reasoning Backwards to Cartel Conspiracies

Seth Stephens Davidowitz’s Everybody Lies argues that internet search trends can help us discover new information about human behavior. Throughout the book, he really struggles with proving the book’s core premise, or at least hedges a lot. In some cases, this is because the example used doesn’t rely on search data (i.e. horse-racing being determined by horse heart/lung size) but in high-reward cases like finance, online dating apps, etc. public data is used and outclassed by private firms, who already have that info and much more. Search data can only make public what someone else knew yesterday.

This book wants to be conspiratorial, and it seems to fizzle out(private information>Google isn’t surprising) but in my opinion it doesn’t go far enough. For instance, SSD points out how movie advertising during sports playoffs depends on which teams lose, but doesn’t suggest that movie theatres or cities might already be trying to predict the Super Bowl (let alone rig it). Is anyone doing that research?!?!?! How accurate is Disney/IMAX at predicting football games?!?! Are they in cahoots with casinos?!?!?! Are sports rigged by Big Cinema? I need to know!

The fact that SSD doesn’t bring any of this up suggests that trying to do anything through public info fundamentally blinds you the truth staring social science research in the face: Either you’re in private knowledge or you’re hopelessly behind.

1 Like

I sort of have a feeling that you are both much more correct than one would naively expect but also much more wrong. First, there are a ton of such conspiracies so it’s very wise to take the possibility of them existing into account but also the space of problems in which conspiracies might arise is way higher than the number of people who are able to take advantage of them so in most cases no such conspiracies exist in fact.