To Turf or Not to Turf, That is the Question [Part 2]: Applications

In the previous video I have provided “data driven” approach in deciding should your team train on the artificial grass or not. We have covered how I made sample data, how we train predictive model using CART trees (regression trees) and caret package in R language, and how to visualize the model performance.

In the video below I am deploying the model to make the decision between 5 variations of weekly plan. The goal is to minimize the morning soreness on the day of the game. We can use our model trained of observational data to help us predict athletes reactions, and hence help us in making optimal decision.

You can easily expand the model/code to make predictions for each athlete separately, rather than on the team level, which could be the next step when deciding optimal strategy for an individual rather than on the team level.

Click Here to Download the Example Data

Related Articles

Predicting Performance Using Rolling Averages

I wrote before on using Banister model to predict performance and injury from training load. In the video below I am talking about using rolling averages to predict performance, as well as explaining differences between supervised and unsupervised learning and model overfit. Members can download the Excel workbook below the video and follow the “exercises” and play with the…


Your email address will not be published. Required fields are marked *

Cancel Membership

Please note that your subscription and membership will be canceled within 24h once we receive your request.