Weird Science: How to Know Your Predictive Discovery Is Not BS


First order of business...

Make sure you fill out our survey if you haven't already. We've only heard from ~10% of our members, so we definitely want to hear from more of you. Averagte completion time is only 4 minutes, so please take a few minutes to fill out this super short survey.

The next order of business...

In celebration of the winter solstice and general holiday-ness of the cold months, we will be hosting a repeat performance by our old friend Eric Siegel.

This event will have high attendance and will also be combining forces with other meetups under the Burlington Code umbrella for a big party afterward. So plan to stay late and grab some drink and food with us at Skinny Pancake and socialize with all the cool folks in the Burlington tech community. We will provide $7 drink vouchers to everyone.

Also, this talk is not just for data scientists (!). Eric is quite expert at explaining and making data science concepts accessible to broad audiences. Please, help expand our community and invite non-data sciencey folks (like your boss) who you think might be interested in what data science has to offer organizations seeking to become more data driven.

How to Know Your Predictive Discovery Is Not BS

An orange used car is least likely to be a lemon." At least that's what was claimed by The Seattle Times, The Huffington Post, The New York Times, NPR, and The Wall Street Journal. However, this discovery has since been debunked as inconclusive. As data gets bigger, so does a common pitfall in the application of standard stats: Testing many predictors means taking many small risks of being fooled by randomness, adding up to one big risk.

In this talk, Eric will cover this issue and provide guidance on tapping data's potential without drawing false conclusions.

Eric founded Predictive Analytics World (, has written a best selling book on prediction ( and speaks widely on how predictive analytics works, and the ways in which data science delivers value to organizations across industry sectors.