Stats Perform: Streamlining Live Capability in Sports Data and Analytics

Stats Perform

Nancy Hensley, Chief Product & Marketing Officer (CPMO), Stats PerformNancy Hensley, Chief Product & Marketing Officer (CPMO)
Live streaming services have drastically altered the sports industry. Given the technological innovations across all aspects of the chain, viewers are embracing the possibilities and flexibility of live streaming while the media and broadcast teams are enjoying a variety of benefits, including global reach and new revenue opportunities. The upswing in data intelligence has further pushed online sports streaming to break in-depth, data-driven stories faster, in addition to using the data insights to improve services and investments around sports.

Driving the technological innovation to transform how sporting events are consumed worldwide is Stats Perform, the sports tech leader in data and AI. The company fuels the power of AI and ML to look into deeper aspects of the game and impact sports around the world by launching new fan engagement, sports betting, and team performance solutions with improved content around sports.
In an interview with CIO Applications, Nancy Hensley, Chief Product & Marketing Officer (CPMO) of Stats Perform, talks about live streaming capabilities and how AI-powered data around sports is used to leverage outcomes to the stakeholders.

Could you chart out a brief overview of the genesis of the company?

Stats Perform is a sports data and AI company. The genesis of the company goes back to the 1980s with Bill James, one of our founders who published the Baseball Abstracts, pushing the envelope on sports analytics by forming Project Scoresheet, a crowdsourced method to collect and distribute data. It marked the early days of what became Stats, and the foundation of today's Stats Perform.

We have been instrumental in pioneering the data revolution in sports by employing live tracking data inside of a game, as well as streaming, to engage sportsbooks and betters. The transition has furthered the data tracking from a camera-based system to computer vision, like AI-powered capabilities known as Qwinn.

Can you give us a glimpse of how Qwinn, the AI capability, works?

Qwinn powers our AI tools and models. It has been used to create over 140 of our models used in applications like Pro Analysis and Pro Recruitment products, which help teams and leagues evaluate their game strategies and optimize recruitment choices. It could also be used to tell a better story or used inside of a fantasy application making predictions for a player or a season or a team.

Most of our unique, patented core models are multifaceted. Depending on the complexity, these stand-alone models can become features that can be fed into other models. For instance, our models can be consumed via an API or a media product like PressBox to help create visualizations and content in real-time.

Currently, we are broadening our focus to live models and how they can be used to improve team performance by assisting the teams in building better strategies and recruiting effectively. We have had remarkable success with teams like Croatia, a small market-budget team then who used one of our Pro Analysis tools to advance to the World Cup in 2018.

Fueling from the success in operating live capabilities, we are designing it to be more consumable in the form of media and broadcast applications.

We've always been on the leading edge of innovation. We were the pioneers to track data in sports, help fuel the explosion of in-play betting as well as develop AI-related data from broadcast with computer vision capability, all because of our focus on innovation

Our newly launched PressBox Live is a live dashboard that helps surface insights into the game and related content through a single user interface. It will provide a competitive edge for broadcast and media to grow viewership and ratings which bring back more advertising and sponsorship dollars. A greater focus on a more content-rich environment could also change the sports betting experience. Modeling natural language generation (NLG) capability through AI has helped convert structured data into human-sounding text, enabling our media clients to cover basic information, like game previews, bios, recaps, etc., at scale in an automated fashion.

What is the marketing strategy for Pressbox, and how does the process change when the clientele changes?

Conducting pre-game research, keeping track of interesting storylines, and providing broadcasters with real-time information that is scattered across multiple platforms with separate logins can get disorienting. PressBox was built as a solution to streamline this capability. It serves as a single login platform, with tools that help surface the relevant insights and information around the sport, separate signal from noise, and view multiple events and games in the same dashboard.

PressBox has built-in onboarding capability, like PressBox Video that walks you through how to create your own video highlights, edit right on the screen, and then download it. There are heat maps and visualizations with relevant in-depth player data as well as updates on live win probability models as the game progresses, all on one screen in PressBox Live. By integrating all this into a single platform, we're trying to make AI extremely consumable for the media team or the viewers -whether it's creating a broadcast or writing a live blog, or managing social media- by channeling useful, relevant, and personalized information in the field of sports.

How does Stats Perform maintain its competitive edge?

We've always been on the leading edge of innovation. We were the pioneers to track data in sports, help fuel the explosion of in-play betting as well as develop AI-related data from broadcast with computer vision capability, all because of our focus on innovation.

Last year, we put together an innovation team in addition to the data science team to quickly incubate and test new concepts without impacting our normal development cycles. The addition of the new team has helped develop great applications without impeding the ongoing research and progress in many areas, like computer vision, contextualized viewing experiences, and keeping a very tight alignment between product engineering and data science. We have created a head start over our competitors in the sports AI field by combining expertise in storytelling and data science, as well as the ability to create, manage and fully productionalize our AI assets.

What does the future hold for your organization?

We believe the arms race in sports AI is to take place through live streaming. We've opted to focus a large portion of our AI capabilities on computer vision as viewers are demanding tailor-made content in real-time, with OTP becoming the preferred viewing experience that gives more contextualized, layered information.

In addition, we are applying live modeling capabilities to help in coaching. For instance, employing AI as the coach's 'eye in the sky' using multiple models like momentum analysis, playing styles, and shape analysis to formulate better patterns and effective modifications to enhance the ability and the information around the game.

We are also trying to elevate the experience of the fans and betters as well as the strategy of the teams with advanced technology. Our recently launched direct-to-consumer experience called the Analyst (www.theanalyst.com) was built to address the ever-increasing demand for fan engagement and data around the game and better understand what is working in Fan Engagement. With The Analyst, we are breaking new ground as a B2B company by giving everyone a taste of how to tell better stories, analyze the game or a player, and do more than write a whitepaper about it. We wanted to build an experience for the fan, and show how data can be consumed through our simulation models, and change the way sporting events are enjoyed across the globe.

Read Also



Featured Vendors