Esports research ideas: Tactical insights – Tournaments
As esports continues to evolve and gain traction within mainstream entertainment, its capacity for data analysis far surpasses that of traditional sports such as football or baseball. Dr. Darina Goldin emphasizes this point, highlighting the transformative potential of data in the competitive gaming arena.
When discussing the best support player in League of Legends, debates can ignite among fans for hours. Yet, the reality remains: without a robust, data-driven approach to evaluate player performance, arriving at a definitive answer is virtually impossible.
Currently, the analysis of player capability in esports often relies on subjective narratives crafted by a select few analysts and the public’s perceptions on platforms like Twitter or Reddit. While these discussions can be entertaining, they are often devoid of scientific rigor.
Understanding player skill is paramount in esports, primarily due to the financial implications; a player’s value can significantly influence salary negotiations. Additionally, it raises questions regarding an individual’s adaptability and overall contribution to team success. Is a player excelling due only to their current team dynamics, or would they thrive in any environment?
The value of discussions surrounding player performance is inherently enhanced when grounded in concrete data. Fans gravitate towards sports like baseball partly because of the rich statistical insights it offers, as popularized by the book “Moneyball.” This depth of analysis is equally applicable to esports, and leveraging such statistics could greatly enrich the discourse.
Contrary to assumptions, esports isn’t lagging behind traditional sports in analytics. Organizations across leagues like the NFL and NBA have recently started employing dedicated analytics teams. A notable example is Premier League player Kevin De Bruyne, who effectively utilized data scientists to demonstrate that he was “significantly underpaid” just a year ago.
Despite some academic exploration into player evaluation, comprehensive research remains sparse. This leaves significant gaps in our understanding of player roles and group dynamics, especially when assessing who among the best midfielders globally truly stands out.
Esports: A Leader in Data Analytics
As enthusiasts of esports, one might typically look to traditional sports for guidance. However, in this realm, it is esports that should lead the way, thanks to the unprecedented volume of data available.
Effective analysis in traditional sports often requires meticulous record-keeping of each play, the use of specialized tracking technology, or intricate video analyses. In contrast, esports generates data autonomously.
Each match unfolds on a server, automatically logging every player action—from positions to item purchases and attack launches—at an exceptionally high temporal resolution. For instance, Riot Games provides licensed data through Bayes Esports Solutions, capturing nearly every aspect of gameplay.
Moreover, capturing micro-level data such as mouse movements and keystrokes enhances our understanding of player response times and focus levels. The clearly defined roles players occupy further simplify this process. For instance, in CS:GO, the designated AWPer is easily identifiable, while roles such as the jungler in League of Legends are explicitly labeled in analytics data.
While traditional sports research often begins with identifying player roles, esports allows us to dive directly into analyzing player performance metrics and skill sets.
However, it’s essential to acknowledge the impact of game patches on performance metrics—a relevant consideration for both research and gameplay. Adapting to patches represents another skill set worth measuring, particularly as players navigate shifts in game balance and mechanics.
Given the wealth of data at our disposal, there is no reason why esports should not spearhead innovative research into team dynamics and performance metrics. Traditional sports should be looking to the esports sector for insight, not the other way around.
Dr. Darina Goldin is the director of data science at Bayes Esports. Her journey began in competitive Team Fortress 2 during her graduate studies. Though she no longer competes, she remains a passionate esports advocate. At Bayes, Dr. Goldin develops predictive models for games like Counter-Strike, Dota 2, and League of Legends. When not immersed in data, she can be found honing her Brazilian Jiu-Jitsu skills at the gym.
