Guest post by Jonathan Taplin, Professor at the University of Southern California and the Director of the school’s Annenberg Innovation Lab. Next week Professor Taplin will be participating in the IBM Information on Demand and Business Analytics Forum (IOD11).
Twitterball – Games 1 & 2
IBM and USC Annenberg Innovation Lab students continue to analyze fan sentiment in Major League Baseball’s Fall Classic. After just two World Series games, the sheer volume of tweets is astounding — almost twice as many as the entire Championship Series. For a World Series involving two small market teams — resulting in lower TV ratings than last year’s series — social media engagement is extremely high.
In terms of number of “sentiment” tweets – that is, tweets both positive and negative, our sampling showed:
– In game 1: The St. Louis Cardinals’ Arthur Rhodes and the Texas Rangers’ Nelson Cruz were the most tweeted among players and coaches.
– In Game 2, the Cardinals’ Jaime Garcia and the Rangers’ Elvis Andrus took the leader crown for sheer numbers of tweets.
But despite receiving the highest number of tweets, these players lagged their teammates and coaches in our ‘T’ scores – the ratio of positive to negative sentiment indicated in tweets — regardless of their performance. In Game 1, Cardinals’ manager Tony La Russa – who has been lauded for his great decision making, came away with 97% positive sentiment, while in Game 2 Elvis Andrus led his squad with 98% positive sentiment. Albert Pujols obtained a 96% positive rating – indicating his role in a key 9th inning play that hurt his team didn’t necessarily hurt his “T-score” — while the Rangers’ Ian Kinsler, who sparked the Rangers’ 9th inning comeback, scored 92%.
First, based on our analysis, one trend seems to be that those with the most face time on TV during the game tend to be those who receive the greatest number of tweets. But that doesn’t necessarily translate to a positive sentiment score. And the fans are taking to social media platforms like Twitter to let the world know who they think are the heroes and zeros of the game.
But beyond baseball, we are learning that this kind of analysis is just the tip of the iceberg of what we can do with analytics and the massive explosion of information coming from millions of individuals in the digital world. Understanding actual fan (or consumer) sentiment – about baseball, movies, fashion trends, the latest mobile phone and more – versus what the so-called experts might predict will be a hot seller or trend, can transform the kinds of make-or-break decisions that large and small businesses are faced with all the time. And it can help companies and people make those decisions before it is too late – like what kinds of promotions to run, whether to sell a new fashion to the masses online or introduce it on Main Street.
The more people share what they actually think about a product, a person, a company online — the more their influence grows. And the more businesses ignore those opinions, the worse off they may well be. That’s why this work between IBM and AIL is so important. The shift in the balance of power from traditional to new media – and from the influence of a few experts to many millions of actual consumers — requires students to gain new skills in analytics, and help more organizations apply new technologies to understand conversations taking place via digital sources to stay competitive.
Surely, there will be some exciting new data trends to appear from the upcoming games. Maybe ‘Twitterball’ could become the next ‘Moneyball.’ We’ll share new insights next week.
In the meantime, let’s get back to the ballpark, or at least to the Twitterverse.
- Read about analysis of Games 3 and 4.
- Read more about “Behind the Diamond: Understanding MLB Fan Sentiment.”
- Learn about how IBM and USC AIL are conducting the social media analysis project.
- Check out images from the World Series Social Sentiment Index