IBM: Eli Manning Upsets Tom Brady in Social Super Bowl

According to an analysis by IBM and the University of Southern California, Eli Manning of the New York Giants has edged out the New England Patriots’ Tom Brady as the top QB in the Super Bowl.

If Eli Manning doesn't beat Tom Brady on the field on Super Bowl Sunday, he has at least thus far won the battle of public sentiment, according to an analysis of Twitter traffic by IBM.

In a joint effort with the University of Southern California's (USC) Annenberg Innovation Lab, IBM used its analytics software to assess public sentiment of the two teams in the Super Bowl, in particular the quarterbacks. IBM and USC have conducted the first sentiment analysis of the two Super Bowl quarterbacks to illustrate how new analytics technologies make it possible to understand positive, negative and neutral sentiments.

IBM and USC analyzed nearly 1 million tweets to find that New York Giants' quarterback Eli Manning held an edge over New England Patriots' quarterback Tom Brady in terms of positive sentiment€”known as "T-Score" in IBM parlance. Manning holds a T-Score of 66, with Brady trailing behind at 61.

However, until Friday, Feb. 3, Brady held a lead over Manning. But Michael Azzi, an IBM spokesman said overnight the numbers turned for Manning in an eight-point shift.

"We think the reason is largely due to two factors," Azzi said. "One is a backlash against all the coverage Eli's brother Peyton is getting this week when he's not even playing in the game. And two is the unusual confidence the Giants and their fans are exhibiting publicly even though they are still a three-point underdog."

Azzi said this shift illustrates very well what happens in today's real-time social media world.

"As in Eli Manning's case, an individual or company's brand can be suddenly boosted or downgraded by external conditions they have no control over€”an acquisition, a competitor's recall, a natural disaster where there suppliers are, etc."

Interestingly, it appears that wide receivers have upstaged the quarterbacks, who are being positioned in the news media as the chief protagonists. The Patriots' Wes Welker is No. 1 in positive sentiment and the Giants' Victor Cruz is a close second. Meanwhile, Peyton Manning's sentiment rating of 63 is close to the two quarterbacks playing in the game, even though he's been out injured all year. And as the coaches go, New York's Tom Coughlin leads New England's Bill Belichick by a T-Score of 76 to 68.

What IBM does with its quickly emerging Smarter Commerce strategy is use analytics to decipher market shifts and the customer sentiment shifts that precede them. The ability to detect these changes as they happen can help a retailer, a bank or a mobile provider adjust its offerings, prices, customer service, even supply chain to meet these mercurial demands.

Using advances in analytics, companies, academics and journalists can gain new insights into consumer perceptions via social media. IBM and USC are using new technologies to distinguish irony and even apply "machine learning" to figure out which tweets are just background noise and which are truly important.

The Super Bowl data that IBM and USC have gathered and analyzed affirms that statistics are certainly an important part of sports. But also understanding social media data can be a powerful tool that impacts the bottom lines for businesses of all sizes in all industries.

The players in the Super Bowl can be viewed as brands€”and many of the more popular ones rely on their "brands" to hawk products for large brand-name companies. Each player carries his own reputation in the market and his own value. Contrary to traditional belief, it is now possible to measure the sentiment around brands€”of people or companies€”in social media like Twitter.

IBM did a previous sentiment analysis October on the World Series and introduced the idea of the 'T Score' or Twitter score. With this Super Bowl project, IBM is taking the brand measurement concept one step further. Analytics now makes it possible to understand positive, negative and neutral sentiments of social media commentary in real time.

More than 100 million people expected to tune in to the Super Bowl broadcast. With more than 800 million Facebook users and 200 million tweets a day, the growing popularity of social networking has also created a social-savvy workforce.

IBM's 2011 CIO Survey of 3,000 global leaders indicated that more than 55 percent of companies identified social networking as having a strategic significance to their company's growth.

Forrester Research estimates that market opportunity for social software is expected to exceed $6 billion by 2016, an increase of 60 percent annually from 2010.

And a recent IBM global study of more than 1,700 chief marketing officers (CMOs) revealed that most of the world€™s top marketing executives recognizes a critical and permanent shift occurring in the way they engage with their customers. This shift requires new marketing approaches, tools and skills to effectively reach customers due to rising online, mobile and social trends.

The USC Annenberg Social Sentiment Index enables students to define areas of research and use IBM Social Analytics technology to explore how it can be used by organizations from news outlets and journalists to movie studios and film marketers to better understand, respond and predict public sentiment. To date, the Index has been applied to film forecasting in order to accurately predict movie blockbuster success rates and most recently was used by students to identify top trends for retailers from the New York Fashion Week shows.