Hottest vs Warmest 100 and the effects of social media

The Australia Day ritual of listening to Triple J's Hottest 100 countdown was this year enhanced by seeing how accurate the Warmest 100 team had been.   For the whole background on how a few geeks used the votes of a small sample published on social media to compile a prediction list the article in The Vine is a good read.

The answer was,  very accurate, but not surprisingly so.  The folks involved published a cool table of their accuracy, which a journo kindly graphed. The simple highlights are that 92 of the songs in the Warmest 100 were in the Hottest 100, they got 8 songs in their correct positions inclusding the top 3 and they correctly predicted the top 10 but not in order.

That this isn't surprising is due to statistical analysis by an Australian economist currently working in the USA. 

The correlation is not surprising, though economist turned politician Andrew Leigh did suggest on Twitter  a possible additional variable "#Warmest100=sharers. #Hottest100=everyone. So extroverts prefer Frank Ocean, The xx, Passion Pit, Grimes..."

There is, however, also another angle to this and that is not so much whether the published choices are an accurate sample, but whether they actually affected the result.  A paper published in 2006 titled Experimental Study of Inequality and Unpredictability in an Artificial Cultural Market studied the effect of social connection on music.  Quoting from the abstract;

Hit songs, books, and movies are many times more successful than average, suggesting that “the best” alternatives are qualitatively different from “the rest”; yet experts routinely fail to predict which products will succeed. We investigated this paradox experimentally, by creating an artificial “music market” in which 14,341 participants downloaded previously unknown songs either with or without knowledge of previous participants' choices. Increasing the strength of social influence increased both inequality and unpredictability of success. Success was also only partly determined by quality: The best songs rarely did poorly, and the worst rarely did well, but any other result was possible.

In brief, the research suggests that the practice of voters sharing their votes online will shape the final result. If that is the case then it is no wonder that the Warmest list was such a good predictor.