Song Ranking based on Piracy in Peer-to-Peer Networks

Music sales are loosing their role as a means for music dissemination but are still used by the music industry for ranking artist success, e.g., in the Billboard Magazine chart. Thus, it was suggested recently to use social networks as an alternative ranking system; a suggestion which is problematic due to the ease of manipulating the list and the difficulty of implementation. In this work we suggest to use logs of queries from peer-to-peer file-sharing systems for ranking song success. We show that the trend and fluctuations of the popularity of a song in the Billboard list have strong correlation (0.89) to the ones in a list built from the P2P network, and that the P2P list has a week advantage over the Billboard list. Namely, music sales are strongly correlated with music piracy.