Real-time Analytics
The PGA TOUR wanted to give fans real-time insight into tournament outcomes, but doing so required a machine learning model that could run thousands of simulations and process nearly four billion records fast enough for live event coverage.
Reported impact
The PGA TOUR can run its model quickly, scale week to week as event complexity changes, and give audiences real-time win and cut probability insights without taking on the operational overhead of managing infrastructure.
The PGA TOUR built its Win, Cut Probability model on AWS to calculate a player’s chances of advancing in and winning a tournament while tournaments are in progress. The solution uses Amazon Elastic Container Service (Amazon ECS) with AWS Fargate to run thousands of simulations without managing infrastructure, then relies on Amazon Redshift and Amazon DynamoDB to support analytical and low-latency application workloads behind the fan experience. This serverless architecture lets the PGA TOUR scale week to week as tournament complexity changes, keep operations lean, and deliver immersive probability insights across multiple viewing platforms.
Tournament and player data feeds the Win, Cut Probability model, which runs thousands of simulations on Amazon ECS with AWS Fargate and serves analytical outputs backed by Amazon Redshift and Amazon DynamoDB across PGA TOUR viewing experiences.
AWS Serverless helps us scale reliably week to week as we move from event to event.
Amazon DynamoDB is a fully managed, serverless, key-value NoSQL database designed to run high-performance applications at any scale.
Each event brings its own set of complexities from size and scale, and what we need to do.
The PGA TOUR introduced Win, Cut Probability, a novel ML-powered analytics model that provides real-time insights into the likelihood of outcomes during tournaments.
Using Amazon Elastic Container Service (ECS) with AWS Fargate to run thousands of simulations and process nearly four billion records, the PGA TOUR can take advantage AWS’s elasticity, run their model quickly and scale as needed, while offloading the operational overhead of infrastructure management to AWS.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
AWS Sample Est. · Dec 18, 2025 Pinterest Pinterest needed to scale AI-powered visual discovery to 600 million monthly active users while keeping the platform a positive, trusted, and safe experience differentiated from competitors who optimize for viral… 10 million AI-powered recommendations per second Amazon Bedrock · Amazon EC2 · Amazon EC2 G5 instances AWS Sample Oct 1, 2024 Swimming Australia To maintain its legacy on the world stage, Swimming Australia sought to transform the vast amounts of raw performance data it had on its athletes into actionable, real-time insights. Amazon QuickSight · Amazon SageMaker AWS Sample Sep 20, 2024 TRACAB TRACAB needed more scalable compute and lower-latency data delivery for computer-vision sports tracking as its features became more complex and its on-premises arena-based setup limited deployment speed and accuracy. average latency of 0.85 seconds Amazon DynamoDB · Amazon EC2 G4 Instances · Amazon EC2 G5 Instances 