This paper explores the use of social media content, specifically Twitter, to predict real-world outcomes, focusing on box-office revenues for movies. The authors demonstrate that a simple model based on the rate of tweets about particular topics can outperform market-based predictors, such as the Hollywood Stock Exchange. They also show that sentiment analysis can enhance the forecasting power of social media data. The study uses a dataset of 2.89 million tweets from 1.2 million users over a three-month period, analyzing pre-release attention, post-release sentiments, and their impact on box-office performance. The results indicate that the rate of tweets is a strong predictor of box-office revenue, and sentiments become more influential after the movie is released. The authors conclude that social media can be a powerful tool for predicting future outcomes and that this method can be extended to other consumer products and topics.This paper explores the use of social media content, specifically Twitter, to predict real-world outcomes, focusing on box-office revenues for movies. The authors demonstrate that a simple model based on the rate of tweets about particular topics can outperform market-based predictors, such as the Hollywood Stock Exchange. They also show that sentiment analysis can enhance the forecasting power of social media data. The study uses a dataset of 2.89 million tweets from 1.2 million users over a three-month period, analyzing pre-release attention, post-release sentiments, and their impact on box-office performance. The results indicate that the rate of tweets is a strong predictor of box-office revenue, and sentiments become more influential after the movie is released. The authors conclude that social media can be a powerful tool for predicting future outcomes and that this method can be extended to other consumer products and topics.