About TL Brand Engagement

The project’s mission is to create an application to measure overall brand engagement for the content posted on Team Liquid’s League of Legends YouTube channel.

Team Chickenjoy built a data science model which combines views, likes, and comments into an overall engagement score.

Looking at views, likes, and comment counts on their own, can lead to confusion and indecision about what content Team Liquid’s YouTube fanbase actually likes. The model’s engagement score presents a simple, easy-to-understand metric for each video. Content producers will now be able to gain insights into what is relevant to Team Liquid’s core League of Legends fanbase.

Why it is special

Platforms such as YouTube do not share data with content creators about customers. It makes it very difficult to understand your core customers. These are the people who come back day after day to watch your videos and sometimes post comments.

The TL Brand Engagement model can be an important tool to estimate the size of Team Liquid’s core League of Legends audience on YouTube. It can be used for each Team Liquid squad’s YouTube channel.

Each major platform is different. Team Liquid fans on Instagram behave differently than fans on YouTube. Not to mention, the way each platform measures engagement is different. The TL Brand Engagement model has the potential to be adapted across other platforms such as Twitter, Instagram, and Facebook. It can give content producers insights into what can be cross-posted across platforms, and what is best for a single platform.

Technical overview

TL Brand Engagement is a multivariate statistical model. The model uses the like and comment counts from each video and creates a proportional metric relative to the view count. The multivariate statistical model takes these metrics and creates an overall engagement score. The overall engagement score is normalized between 0 and 1 to simplify its interpretation.

The application is an interactive web app to visualize the overall engagement score for each video.

The data used for the model and its application prototype is eleven of the most recent videos on Team Liquid’s League of Legends YouTube channel as of December 3, 2022.

Model pros & cons

Data science models have their pros and cons, and TL Brand Engagement is no different.

Pros: Allows you to measure which content is engaging to your core audience.

Cons: Popular content with high view counts have broader audiences outside your core audience and tends to have lower overall engagement. Team Chickenjoy addresses this by identifying the YouTube videos which fall into the top 90 percentile view count range. The results are presented as a table in the application.


We are a father-son team.

Jay “Chickenjoy” Campanell (son, LoL rank: Diamond II). Jay is a high school senior. He is interested in a career in the e-sports industry. He is seeking to get started with data science internships. His goal is to have five years of experience in the e-sports industry by the time he graduates from college.

Rob “st4rtl” Campanell (father, LoL rank: Iron II). Rob is a retail industry data scientist. He is an avid e-sports fan who watches LCS, LPL, LCK, VCS, and PCS. He has aspirations to become a Valorant Game Changers team owner.

Let's Go Liquid

We became Team Liquid fans after seeing co-owner, Steve Arhancet, speak at SXSW in 2018. He is an articulate, well-spoken e-sports visionary. He has the same passion and love for the game as we do. We threw aside the “Paid by Steve” and those 4th Place thing memes and started cheering “Let’s Go Liquid”.

We also want to thank Team Liquid’s brand sponsor, Verizon for their activation at Worlds 2022 Finals where we got to meet Team Liquid pros, BWipo and Bjergsen.

Built With

  • api
  • r
  • shiny
Share this project: