Why Team Analytics?

Team Analytics was born at Nova School of Business and Economics, after the realization that students often want to learn all about data analytics, but few achieve mastery. Team Analytics was developed to increase engagement, and it does exactly that, every year. In Team Analytics students engage with data for prediction, experimentation, and optimization in a competitive environment. This means that class concepts find immediate applicability in Team Analytics. The result? Students love the experience, and engage like nothing else.

The main design principle of Team Analytics is to emulate the dynamic environment of startup companies, where the primary task is to understand market dynamics and adapt accordingly. In Team Analytics, students compete by analyzing data to optimize their car’s performance and strategize for races. The simulation emphasizes data-driven decision-making as the key to success. Behind the scenes, a carefully-crafted model ensures that the quality of the students’ decisions maps to outcomes, ensuring a fair context as well as a signal-to-noise relationship that is ideal for learning from good and bad decisions.

Two levels: Basic and Advanced

Students form teams and step into the roles of racing strategists and data analysts. The simulation features two levels of difficulty, allowing the instructor to cater to the level of the course. The basic level is suited high school level, whereas the advanced level will benefit from the most advanced techniques, such as sophisticated prediction models, optimization, variable selection, etc.

In a nutshell

The simulation is divided into race sessions. In reach race, teams analyze training data to map how track characteristics and weather conditions, together with the car setup, translate to lap times. Teams then prepare for the race by deciding on the ideal car setup. They can also experiment by taking the car around the track. This allows them to devise the race strategy: When to pit and which tyres to equip. During the live race, students learn from their rivals’ mistakes as well as their own (if any!), and use the feedback to improve.

Working with data

Accessing data for analysis couldn’t be easier. Students can access their team’s data by downloading it from the simulation, or directly from their favorite analytics software. MATLAB^®, Python and RapidMiner are supported.

Easy course integration

Instructor are provided with materials to introduce the simulation and connect it to class concepts. Throughout the simulation, the instructor has access to all student views and is able to monitor championships in detail.

Great signal-to-noise ratio

The training data is modeled so students face a reasonable (but not insurmountable) challenge. In order to ensure students get the maximum feedback, teams compete in the same championship without strategic interaction, meaning each team’s performance is solely determined by their own strategic decisions and data analysis skills. This enables students and instructors to accurately evaluate team performance in a low-noise environment, and gain valuable insights into their capabilities.