The project aims to discover a significant impact of social media posts addressed to NBA players before matches with respect to their influence on these players’ game performance. For this purpose, we consider NBA players that are highly active on Twitter and extract tweets that are addressed to them within a short period of time before matches via the Twitter API. A sentiment analysis indicates the attitude of the posts and test if there is a correlation between social media posts and players’ on-court performance. We also try to predict the players’ game performance using a predictive model.
With the growing presence of social media in all areas of life, allowing people from around the world to react to current events in real-time, an increasingly controversial discussion can be noticed. Today more than ever, public figures are exposed to the reactions of millions of people, observing and commenting on every step in their life that becomes public.
Sports athletes, who use social media not only to communicate with peers and fans but also to promote themselves, are no exception to these issues [1]. Among researchers in the sports field, there is a consensus that the mental state of an athlete can have a significant impact on his or her performance [2]. However, only little research has been conducted in order to analyze if and how social media usage of athletes directly influences their performance. Xu and Yu [2] tried to capture the mood of basketball players in the NBA from the tweets they posted just before a match, using sentiment analysis, to analyze how the predicted mood influenced their performance on court. Gruettner, Vitisvorakarn and Wambsganss [3] used a similar approach on tennis players and additionally analyzed the relationship between the number of tweets they posted before matches and their performance within the match. Ott and Puymbroeck [1] list cases where athletic performance appeared to be immediately influenced by the media and conclude that the media has the potential to change the performance of an athlete in a negative as well as positive way. In this analysis we aim to assess this relationship more closely by analyzing how social media posts addressed to NBA players affect their game performance.