Total Instagram Followers
Followers Growth in the Last Year
Number of Posts in the Last Year
Most Popular Post in the Last Year
Engagement Rate
The ‘Which tennis players are winning on Instagram?’ ranking includes the ATP and WTA rankings for both male and female tennis players, respectively.
Players Selection
Based on top 10 ATP & WTA Rankings June 14, 2021.
Factors and Scoring
In order to assess which tennis stars are winning on Instagram, five factors were examined and a score was calculated for each one.
Total Instagram Followers
Number of followers in an account in June 2021 - to assess the players’ popularity with fans on Instagram.
Followers Growth in the Last Year
Number of followers added to an account since June 2020 - to understand which players are gaining or losing popularity with fans over the last year.
Number of Posts in the Last Year
Number of posts added to an account since June 2020 - to illustrate how active the tennis players are on social media.
Most Popular Post in the Last Year
Number of likes for the most popular post in an account since June 2020 - to see which tennis players are producing the most ‘likeable’ content.
Engagement Rate
The average rate of followers who engaged with posts (likes, comments) in an account in the last 180 days - to understand how fans are engaging with each of the tennis players.
The variables “Total Instagram Followers”, “Followers Growth in the Last Year” and “Most Popular Post in the Last Year” suffered from an extreme range. To compensate for this, the data was transformed using a logarithmic scale. As some observations were negative, a constant, in the absolute value of the minimal observation in each variable plus 1, was added to each observation. The formula for this was as follows: = log(x + | min { min {X} -1 , 0 } | ).
Then, all the variables were normalized using min-max normalization. For every variable, the minimum value was transformed into a 0, while the maximal value was transformed into a 1, all other values were transformed into a decimal between 0 and 1. The formula for the normalization was = x – min{X} / max{X} - min(X).
At this point, the values of all variables were added together to create an aggregated score between 0 and 5. To ease the interpretation, the final results were scaled to the range between 50-100 points as follows = X*10+50.
