AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Effect Size of Anomalies on the Bell Curve of Data
When we put all the data together for 17 years and 500 well characterized formal experiments, that's the bottom line. We about two thirds of the events show the effect we predict. But three quarters of them do. And that's not likely if you've got 500 things,. That's like what do you think is going to be the average if you're flipping a coin 500 times? You won't get usually anything like two thirds of them. The overall effect size is in terms of standard deviations - how far out you are in the bell curve of data.