[Paper Review] Indian Premier League (IPL), Cricket, Online Social Media

Megha Arora, Raghav Gupta+1 more|arXiv (Cornell University)|May 20, 2014

Public Relations and Crisis Communication17 references3 citations

TL;DR

This study analyzes social media activity during the 2013 Indian Premier League (IPL) across Twitter and Facebook, using 2.6 million tweets and 774k Facebook posts to examine temporal trends, geographic distribution, sentiment, and the impact of the spot-fixing scandal. It reveals strong correlations between team brand value and online discussion volume, with decay patterns on both platforms following similar power-law dynamics, and shows that public sentiment remained largely positive and unaffected by scandals, indicating IPL's role as primarily entertainment.

ABSTRACT

In recent past online social media has played a pivotal role in sharing of information and opinions on real time events. Events in physical space are reflected in digital world through online social networks. Event based studies for such content have been widely done on Twitter in the computer science community. In this report, we performed a detailed analysis of a sports event called the Indian Premier League (IPL'13) for both Facebook and Twitter. IPL is the most popular cricket league in India with players from across the world. We analysed more than 2.6 million tweets and 700 thousand Facebook posts for temporal activity, text quality, geography of users and the spot-fixing scandal which came up during the league. We were able to draw strong correlations between the brand value of teams and how much they were talked about on social media across Facebook and Twitter. Analysis of geo-tagged data showed major activity from metropolitan suburbs however activity was not restricted to the regions geographically associated with each team. We present a decay calculation methodology, using which we derive that activity died down on both Twitter and Facebook in a very similar manner. Such analysis can be used to model events and study their penetration in social networks. We analysed text for spot-fixing and found that user response to allegations about matches being fixed was cold. The complete analysis presented in this report, can be particularly useful for studying events involving crisis or events of political importance having similar structure.

Motivation & Objective

• To understand the dynamics of social media engagement during a large-scale, scheduled sports event like the IPL.
• To compare user behavior, content volume, and sentiment across Twitter and Facebook during the IPL 2013 season.
• To investigate the impact of the spot-fixing scandal on public sentiment and online discourse.
• To model the decay of event-related activity in online social networks using a decay factor methodology.
• To explore the geographic distribution of social media activity and its alignment with team affiliations.

Proposed method

• Collected and analyzed 2.6 million tweets and 774,186 Facebook posts from unique users during the IPL 2013 season.
• Applied temporal analysis to track hourly activity spikes corresponding to match schedules and major events.
• Used geo-tagged data to map user locations and assess regional engagement patterns, focusing on metropolitan and industrial zones.
• Developed a decay calculation methodology using a 99% threshold to determine when activity becomes insignificant, computing net decay factors for both platforms.
• Conducted sentiment analysis using the LIWC (Linguistic Inquiry and Word Count) tool to classify emotions, social words, and money-related terms in text.
• Performed manual content review of spot-fixing-related posts to contextualize quantitative sentiment findings.

Experimental results

Research questions

• RQ1 How does the volume and timing of social media activity on Twitter and Facebook correlate with scheduled IPL matches and events?
• RQ2 To what extent do team brand value and fan base size predict online discussion volume on social media platforms?
• RQ3 How did the spot-fixing scandal affect public sentiment and emotional tone in social media discussions?
• RQ4 What is the temporal decay pattern of social media activity following the conclusion of a major sports event like the IPL?
• RQ5 How does the geographic distribution of social media users relate to team affiliations and regional demographics?

Key findings

• The net decay factor for Facebook activity was 0.1942, and for Twitter it was 0.2022, indicating very similar decay dynamics across both platforms.
• There was a strong correlation (0.91) between team brand value and online discussion volume on Facebook, and a high correlation (0.79) on Twitter.
• 75.53% of geo-tagged tweets and 89.85% of geo-tagged Facebook posts originated from India, with the highest activity from metropolitan suburbs and industrial belts.
• Despite the spot-fixing scandal, sentiment analysis showed only 0.33% and 0.31% of text content on Twitter and Facebook contained anxiety-related words, indicating low emotional distress.
• Positive emotion words were more prevalent than negative ones, and money-related terms were underrepresented in general, though slightly elevated in spot-fixing-related content.
• Manual content analysis revealed that users largely treated the IPL as entertainment, remaining indifferent to scandal allegations, which explains the low anxiety levels.