Fanlytics

Inspiration

Our project was born from the curiosity of how top NBA players can sway the financial markets through their influence. We were inspired by instances like when Cristiano Ronaldo made critical comments about Coca-Cola, leading to a noticeable dip in their stock. This phenomenon sparked our interest in exploring the dynamic relationship between sports, finance, and public sentiment. With the rapid pace of NBA data and the increasing integration of financial metrics into sports analysis, we aimed to uncover how a superstar's performance or off-court remarks can directly impact sponsor stocks and market trends.

What it does

Our application bridges the gap between NBA player performance, public sentiment, and financial markets by tracking real-time NBA statistics and social media reactions. It analyzes public sentiment using AWS Comprehend (or a custom NLP model) and correlates this data with the stock performance of players' sponsors or affiliated companies. By collecting live game data and analyzing social media mentions, the system predicts potential stock fluctuations based on player actions, whether it’s a game-winning shot or an off-court controversy. This dynamic analysis empowers sports analysts, investors, and fans to understand how athletic performance and public sentiment can directly influence financial markets, offering a unique perspective on the intersection of sports and finance.

How we built it

We built Fanlytics using a robust tech stack that seamlessly integrates front-end, back-end, and cloud services to deliver real-time insights. The front-end was developed using React, providing a dynamic and interactive user interface for visualizing player stats, sentiment analysis, and stock market trends. On the back-end, we leveraged Python for data processing and API integrations, connecting to the NBA API for live player statistics and the Yahoo Finance API for real-time stock data. We utilized AWS Lambda for serverless computing, API Gateway for managing API requests, and AWS Comprehend for natural language processing and sentiment analysis. Data is stored in MongoDB Atlas, allowing for flexible and scalable database management, while Amazon S3 is used for storing datasets, logs, and model artifacts. This combination of technologies enabled us to create an end-to-end system that efficiently processes real-time data, analyzes sentiment, and correlates it with financial market movements.

Challenges we ran into

One of the biggest challenges we faced was integrating multiple real-time data sources, such as NBA statistics, social media sentiment, and live stock market data, into a cohesive system. Ensuring that the data pipelines worked seamlessly while maintaining low latency was a technical hurdle, especially when dealing with rate limits and API inconsistencies. Another major challenge was fine-tuning the sentiment analysis. While AWS Comprehend provided a solid foundation, accurately interpreting the nuanced context of tweets—such as sarcasm or slang—proved difficult. Additionally, setting up AWS Lambda with external libraries like NumPy and managing cross-service permissions created deployment complexities that required multiple iterations to resolve.

Accomplishments that we're proud of

We’re incredibly proud of building an end-to-end system that successfully connects sports performance with financial market analysis. The seamless integration of real-time NBA data, social media sentiment, and stock market trends showcases the depth and flexibility of the application. One of our standout achievements was designing a dynamic sentiment analysis pipeline that can assess how player actions influence sponsor stocks in near real-time. Overcoming technical hurdles, like setting up AWS Lambda functions with complex dependencies and establishing reliable data flows, was a significant accomplishment. Seeing our initial concept transform into a functioning product that offers unique insights into the sports-finance relationship is something we’re truly proud of.

What we learned

Throughout this project, we deepened our understanding of building scalable cloud architectures using AWS services like Lambda, API Gateway, and Comprehend. We learned valuable lessons in handling real-time data, particularly the challenges of synchronizing multiple data streams and maintaining system reliability under fluctuating loads. The complexities of sentiment analysis also taught us the importance of context in natural language processing, especially when dealing with informal social media language. Additionally, we gained hands-on experience in managing permissions and API integrations, as well as optimizing AWS Lambda deployments to handle external libraries and larger datasets efficiently.

What's next for Fanlytics

Looking ahead, we plan to expand Fanlytics by incorporating more advanced machine learning models, such as fine-tuned BERT models, to improve sentiment accuracy and better capture social media nuances. We also aim to broaden the scope beyond the NBA, integrating data from other major leagues like the NFL and MLB to analyze cross-sport market impacts. Another future goal is to introduce predictive analytics that can forecast stock movements based on upcoming games or trending player news. Finally, we envision developing a user-facing dashboard where fans, analysts, and investors can visualize the real-time connections between sports performance, public sentiment, and financial trends, making data-driven insights accessible to everyone