Team Liquid: Capitalizing on unlimited access to team fight data to win the next League of Legends match
Read how world-renowned esports organization Team Liquid is using a machine learning model enabled by SAP solutions to efficiently identify team fights within game timelines and discover new insights for game decision-making. Download the DocumentINTERNAL – SAP Only (delete if public)
Team Liquid: Capitalizing on unlimited
access to team fight data to win the
next League of Legends match
Team Liquid is a world-renowned professional gaming organization established in
2000. The company is one of the most successful esports organizations in the world,
winning over US$49 million in prize money and 140 major titles. Every day, more than
190 championship-caliber athletes in over 20 game titles represent Team Liquid globally.
As a game, League of Legends is very data intense and complex. Analyzing the behavior
and strategies during team fight situations is an essential part of match preparation. But
finding team fights in games that last for 45 minutes was a manual and time-consuming
task, limiting the number of matches Team Liquid could analyze during short breaks in a
tournament. Seeking another way, Team Liquid wanted to find out if the tedious task of
detecting team fights could be automated.
Picture Credit | Team Liquid, Santa Monica, California. Used with permission.
Team Liquid: Capitalizing on unlimited
access to team fight data to win the
next League of Legends match
Team Liquid is a world-renowned professional gaming organization established in
2000. The company is one of the most successful esports organizations in the world,
winning over US$49 million in prize money and 140 major titles. Every day, more than
190 championship-caliber athletes in over 20 game titles represent Team Liquid globally.
As a game, League of Legends is very data intense and complex. Analyzing the behavior
and strategies during team fight situations is an essential part of match preparation. But
finding team fights in games that last for 45 minutes was a manual and time-consuming
task, limiting the number of matches Team Liquid could analyze during short breaks in a
tournament. Seeking another way, Team Liquid wanted to find out if the tedious task of
detecting team fights could be automated.
Picture Credit | Team Liquid, Santa Monica, California. Used with permission.
INTERNAL – SAP Only (delete if public)
“With the advent of our team fight detection model we’re
able to leverage vast amounts of game data stored in
SAP HANA Cloud to discover new insights for game
decision-making. This entirely new line of analysis and
insights for League of Legends is possible only through
this partnership.”
Jesse Hart, Senior Director of Sports Science & Analytics, Team Liquid
Team Liquid
Santa Monica, California
www.teamliquid.com
Industry
Media, sports, and
entertainment
Employees
>300
Products and services
Professional esports
teams
Featured solutions
SAP BTP, SAP Business AI,
and SAP HANA Cloud
>5,000
Working hours per year
saved
>US$120,000
Projected to be saved by
eliminating manual analysis hours
Before: Challenges and opportunities
• Highly dynamic nature of League of Legends, with five players on each team engaging in various
actions at different locations simultaneously, making detecting team fights a labor-intensive process
• Reliance on manual efforts to identify key moments of combat
• High volume of professional matches—especially in tournament settings where teams have limited
preparation time between matches
Why SAP
Capability to develop a machine learning model for automated team fight detection using high-
frequency data from approximately 15,000 esports matches enabled by SAP Business Technology
Platform (SAP BTP), SAP Business AI, and the SAP HANA Cloud database
After: Value-driven results
• Enabled analysts to efficiently identify team fights within game timelines
• Empowered analysts with precise timestamps for each detected team fight, along with information on
participating players, eliminating the need to scan through raw match data
• Significantly reduced the time needed for match preparation and strategic analysis
• Enhanced decision-making and streamlined decision tree design for team fights, helping improve
competitive performance
Automating match analysis with a machine learning model using
SAP BTP, SAP Business AI, and SAP HANA Cloud
96737enUS (25/03) © 2025 SAP SE or an SAP affiliate company. sap.com/terms-of-use
“With the advent of our team fight detection model we’re
able to leverage vast amounts of game data stored in
SAP HANA Cloud to discover new insights for game
decision-making. This entirely new line of analysis and
insights for League of Legends is possible only through
this partnership.”
Jesse Hart, Senior Director of Sports Science & Analytics, Team Liquid
Team Liquid
Santa Monica, California
www.teamliquid.com
Industry
Media, sports, and
entertainment
Employees
>300
Products and services
Professional esports
teams
Featured solutions
SAP BTP, SAP Business AI,
and SAP HANA Cloud
>5,000
Working hours per year
saved
>US$120,000
Projected to be saved by
eliminating manual analysis hours
Before: Challenges and opportunities
• Highly dynamic nature of League of Legends, with five players on each team engaging in various
actions at different locations simultaneously, making detecting team fights a labor-intensive process
• Reliance on manual efforts to identify key moments of combat
• High volume of professional matches—especially in tournament settings where teams have limited
preparation time between matches
Why SAP
Capability to develop a machine learning model for automated team fight detection using high-
frequency data from approximately 15,000 esports matches enabled by SAP Business Technology
Platform (SAP BTP), SAP Business AI, and the SAP HANA Cloud database
After: Value-driven results
• Enabled analysts to efficiently identify team fights within game timelines
• Empowered analysts with precise timestamps for each detected team fight, along with information on
participating players, eliminating the need to scan through raw match data
• Significantly reduced the time needed for match preparation and strategic analysis
• Enhanced decision-making and streamlined decision tree design for team fights, helping improve
competitive performance
Automating match analysis with a machine learning model using
SAP BTP, SAP Business AI, and SAP HANA Cloud
96737enUS (25/03) © 2025 SAP SE or an SAP affiliate company. sap.com/terms-of-use