The HLTV Rating is the most reliable measure of a player's performance in the video game Counter-Strike: Global Offensive (CS:GO). The HLTV rating is calculated using a number of statistics (kills, deaths, average damage per round, etc) but the exact formula or weightages are not disclosed to the public. I use these statistics to predict a player's HLTV rating by fitting a gradient boosting model, and by computing the weightage the model gave to each statistic, I deduce the relative importance of each statistical factor in determining the HLTV rating and analyse the results.
