Institute For The Future

Where's the math?

At IFTF, we are always interested in hearing about new methods for analyzing patterns in human behavior, so this story fits right in with our forecasts about Math World and simulation. But when you come across quantitative claims about the power of a new mathematical method, it's crucial to know:

1. What specifically is being measured/what is determining "accuracy"? 2. Is (1) a reasonable measure? Does it accord with commons sense?

3. What other measurements of accuracy are we missing by choosing (1)?

 

I heard through multiple channels about Malcolm Gladwell (Tipping Point, Blink) interviewing Mike McCready, CEO of Platinum Blue, a company that analyzes songs to attach a number to their potential popularity. McCready throws out a lot of percentages and claims 80-85% accuracy in predicting "hits", 4 times better than record labels that try to pick the single from an album. But he does a poor job in the video and on his website of explaining how they measure songs, how they qualify what a "hit" is, how they measure their success, and if their measure of success is the same as how they are measuring record labels' success without their asssistance.

 

Recent research has shown strong evidence that most hit songs conform to a limited number of mathematical patterns. These patterns cannot be detected by the human ear much in the same way the doctor cannot be see with the naked eye that which the medical Xray reveals.

 

Q: What is the accuracy of Music Xrayâ„¢?

 

A: Most singles released by music labels sound and feel like hits but lack the optimal mathematical patterns. This is a main reason why the industry has less than a 20% success rate. Less than one in 5 songs that are released as singles and promoted with a significant budget actually reach the charts. By using P