HUDstats & SIS: Pioneering AI-Driven Data Technology for Better Live Trading Performance in Competitive Gaming

Project Overview 

We developed an automated data collection pipeline for market-leading betting and gaming company Sports Information Service (SIS), with the purpose of improving their real-time live trading product. 

Background 

Sports Information Services (SIS) is a leading supplier of 24/7 trading services for both retail and online operators globally. SIS offers an end-to-end solution comprising live pictures, data, and on-screen graphics with betting triggers to drive betting revenues for its clients.  

Objectives

As the esports betting landscape evolved, and more opportunities prevented themselves, SIS recognized the need for a more advanced, real-time data collection and distribution solution to improve the quality of their betting product and maintain their competitive edge. 

Objective 1: Improve speed and accuracy of data distribution for the creation of enhanced and more efficient betting products

• Objective 1.1: Reduce the data processing time from seconds to sub-second
• Objective 1.2: Achieve 100% accuracy of data 

Objective 2: Set up the foundation for scalable 24/7 trading operations by increasing the number of esoccer and ebasketball streams on-demand. 

Challenges 

• SIS relied on traditional manual data collection, which is slow and prone to human errors, and is not easily scalable for new content
• Maintaining the processes required constant HR and management efforts to hire, train, retain and manage a considerable workforce
• Limited flexibility in scaling the product on demand, adapting to new game titles or increasing the granularity of data 
• Scaling the process would include sustained HR and management efforts for hiring, training and retaining more staff

Solution 

We designed and developed an automated data collection and distribution pipeline, integrating AI and machine learning technology for extracting, processing, and distributing game data in real time, at a sub-second speed. 

Main Features: 

• Advanced data collection – collecting, processing and distributing over 1,000 data points for every single minute of the game, and reporting on more than 20 major in-game events for esoccer and 40 for ebasketball
• Real-time accuracy - the system enables SIS to offer real-time updates for their betting products with 100% accuracy
• Scalable and efficient - the system can simultaneously manage multiple streams across various esports titles at the same time 
• Data reporting and audit - weekly reports and audit on data quality and incidents 
• Video backups - all streams are backed up in case of operational issues
• 24/7 reporting and alerting - alerting system in place, providing 24/7 support 
• Live data and monitoring dashboard 

Impact 

• Improved speed of distribution: After the incident is captured, data is delivered in less than 0.5 seconds
• Improved accuracy: Attained a 100% accuracy rate in data collection, significantly reducing the margin of error compared to traditional data collection methods 
• System scalability: The system’s scalability facilitated the addition of multiple esoccer and ebasketball streams to SIS’s product offering
• Constant improvement: Constantly we’re training the algorithm to detect new in-game events, thus improving the performance of SIS’s betting products 

Conclusion 

We believe the partnership with SIS represents a significant advancement in gaming live trading, with AI technology playing a crucial role in enhancing the performance of SIS’s product offering. The collaboration sets a new standard in the industry, promising almost unlimited scaling possibilities and developments for the future.