Using AI to Accelerate Competitive Sport

You can also read this report in the latest issue of AI and Technology Magazine. 

The AI industry hears a lot about how AI is accelerating business competition, operations and people’s every day tasks – but perhaps the most entertaining and exciting application of AI innovation is in the sports sector. 

Miami Marlins, the two time World Series winning Major League Baseball team, are deploying AI and machine learning (ML) technologies across their baseball operations. From player performance analysis to fan experience optimization, the Marlins are using AI as part of an organizational rebuild aimed at bringing a World Series championship back to South Florida. 

This rebuild is key to the Marlins’ journey in creating a winning culture and team from the ground up, both on the player’s side and in the front office. This technology-led rebuild began with a strategic front office transformation aimed at creating long-term competitive advantages through data. Leading this development is Brian Chase, Vice President of Baseball Systems, who joined the organization as part of this transformation for the 2025 season as part of a broader front office transformation designed to create sustainable competitive advantages through data-driven decision making. He is now focused on AI to enhance how the team works together to win. 

The Marlins utilise AI and technology “in almost everything we do on the baseball side,” he says. “Every day, our mission at the Marlins is simple: get better players and get players better. “We use data from every pitch, swing and play to understand who our players are today – and who they could become tomorrow.”

At loanDepot park in Miami and the Marlins’ Minor League facilities in Florida, Wisconsin and the Dominican Republic, the Marlins employ vendors including Hawkeye, KinaTrax and Trackman to capture high frame rate video footage during every home game. These systems track all on-field activity, including player movements, batting mechanics and pitching deliveries – generating comprehensive datasets that feed into the organization’s analytical infrastructure.

What’s impressive about the high frame rate technology, is that it enables the capture of granular performance data that traditional observation methods cannot provide. This information is processed through internal database systems and transformed using ML models to generate actionable insights or coaching staff and player development personnel. 

Brian explains that the organization has established “constant feedback loops with coaches, players and staff to give them real time information that they can use to help make better decisions.”

Furthermore, the technology infrastructure supports the franchise’s ultimate objective of winning baseball games whilst developing players more effectively than conventional methods alone.

Now, the Marlins focus extensively on making analytical outputs accessible and actionable for players and coaches. “While we utilise Al on top of our advanced datasets and analytics to come to these conclusions, if we can’t turn those insights into something clear and actionable for a player or a coach, it’s going to be extremely hard for them to integrate those insights into their daily routines,” Brian notes.

The organization particularly employs Gen Al tools to assist coaching staff and scouts in producing detailed reports and analytical content. From Gen Al, scouts observing tryouts at the Marlins’ Dominican Republic facility can utilize Al-powered systems to document player performance during drills, simulated games and practice sessions. 

These Al-generated reports serve as inputs for traditional ML and data science activities conducted internally – and the approach combines qualitative expertise from experienced baseball personnel with quantitative analysis from data scientists and research analysts. Brian describes the methodology as bridging “the qualitative and the quantitative to make really informative decisions.” This means that from Al’s capabilities, the Marlins have cultivated a culture that values both traditional baseball knowledge and advanced analytics, with coaching staff and player development personnel actively seeking data-driven insights to enhance their effectiveness.

The Marlins additionally employ supervised learning techniques to identify factors that correlate most strongly with player performance. Data scientists within the organization build predictive models designed to forecast player development trajectories over three to five year periods, informing decisions about team construction, free agency acquisitions and trade opportunities.  

Brian explains that these models help the organization answer critical questions: “If we acquire, sign or draft a particular player today, how confident and to what degree can we predict that they will become a key contributor for the team in the future?” 

The combination of these projects along with qualitative analysis on player video and live practices, helps us with everything from team construction and free agency decisions to drafting future Marlins. 

“Our goal is to get the best players we can and put the best possible product on the field for our fans.” The predictive capabilities become particularly important for an organization operating with financial constraints relative to some competitors – the Marlins must maximize the value of their player investments through superior evaluation and development processes rather than simply outspending rival franchises.

The Marlins work with more than 40 technology vendors to support their baseball operations, particular standouts being Google Cloud and Posit. Beyond the tracking systems used during games, the organization leverages Google Cloud services. This infrastructure supports the platforming and productionalization of data collected from various sources, as well as the modeling activities and internal applications used across baseball operations. 

Additionally, the franchise has developed a particularly close relationship with Posit, which provides Al and ML platform capabilities. The Marlins use Posit Connect to scale pre-game and post-game reports used across baseball operations whilst building customised applications for players and coaches. 

The team also uses Posit Package Manager to manage their internal R and Python libraries, ensuring consistent deployment of models and analyses across departments. 

“The ability to prototype, deploy and scale applications quickly has made a huge impact on our workflows,” Brian adds. 

Whilst player development is the primary focus of the Marlins’ Al initiatives, the organization also recognises significant opportunities to apply similar technologies to business operations and fan enjoyment.

The Marlins maintain a business analytics division tasked with understanding fan preferences and optimising the customer experience. Potential applications include enhanced security systems to expedite stadium entry, dynamic pricing models for tickets and concessions and predictive analytics for fan attendance and customer lifetime value calculations. 

“We see an opportunity to commercialise some of our baseball data into fan-facing stats and tools,” says Brian. “It’s about helping fans feel more connected to what’s happening on the field.”

Whilst Al has taken the team’s performance to new heights, the proliferation of data sources across professional sports creates both opportunities and challenges for organizations seeking to implement comprehensive AI strategies.

One such challenge is that the Marlins must integrate information from traditional tracking systems with biomechanical data from wearables and qualitative assessments from scouts and coaches. Brian describes this as “bringing all these disparate data sets together to streamline our inputs to our AI activities.”

As a result, the organization combines quantitative analysis with qualitative insights from baseball personnel to create what he terms “a 360 degree loop of building out models, seeing how much correlation there is for certain variables within those models and seeing how real life  events play out – and the cycle continues.”

This iterative approach treats each practice session and game as an experimental opportunity to refine predictive models and validate analytical conclusions against real-world outcomes.

As AI capabilities continue advancing, Brian anticipates significant opportunities for sports organizations to develop unique competitive advantages through technology implementation. The convergence of increasing data availability, improved analytical tools and growing acceptance of technology-driven insights alone creates conditions for sustained innovation in player development and team operations.

Therefore, the Marlins view their current technological investments as foundational elements of a long-term strategy to compete effectively despite financial constraints. “MLB teams are all competing at the end of the day, and the goal for the Marlins each day is to go out and beat that other team across from us on the field and provide an entertaining product for our fans,” Brian says.

It seems that part of what makes the Marlins unique is their approach emphasizing sustainable system development that can support decision-making processes across multiple seasons and player development cycles.

Over the next 12 to 18 months, the Marlins plan to continue building technological infrastructure designed to maximise insights and analysis capabilities. “From a technological point of view, we are really working to build long-term sustainable systems that we can use for getting the most insights and analysis out of our data,” Brian concludes.

“This will help us make the best possible decisions to bring the best players to the field and also provide a winning culture and an environment to help us bring a third World Series championship to South Florida.”