‘Do songs tell stories?’ An empirical analysis of the effect of emotional arcs on success in a national song contest

Abstract This article explores whether the lyrics of songs influence their success in a national song contest. We constructed emotional arcs from over 6.763 song lyrics and examined their impact on contest performance. Our findings reveal that these emotional arcs echo established behavioral phenomena such as the ‘paradox of negative emotion’ and the ‘peak-end rule’. Typically, songs characterized by sorrowful plots and endings receive higher scores than those with more joyful emotional themes. Furthermore, such songs are more likely to be awarded points initially. Despite this, contestants often choose songs with positive endings for their performances. To isolate the effects of lyrics from other song characteristics, we utilized Spotify metrics, including danceability, energy, and tempo. This methodology allows us to differentiate the influence of musical and lyrical elements on listener decision-making. This study contributes to the growing literature on the behavioral aspects of music and its impact on decision-making.