Creating a Poisson-model to determine the optimal strategy for getting Kudos on Strava

Today I will continue looking at Strava data (see my previous post: https://www.andrewaage.com/post/analyzing-strava-data-using-r/). This time, the goal is to create a Poisson-model to determine how to become popular on Strava, answering vital questions such as: What time of the day should I post my training rides? What is more important, distance or speed? Should I include pictures on my activity? Which day of the week should I ride?