STA report:How to increase the win rate of league of legends in rank - Statistics Project Example

Players need to take enemy base down to win. Usually, each player needs to kill minions or enemy to earn extra gold to buy more powerful items. The more kill you have, the chance of win would be higher. Nonetheless, in this research, the focus is on how to increase the win rate of league of legends in rank. Most definitely, in such a game, there are a number of variables which dictate the win rate. Tow variables, namely, KDA and PKA. KDA takes into consideration three important variables, names, kills; number of enemies killed, assists: number of assistances one gets, and kills; number of enemies killed.

The paper evaluates for possible trends in the variables that are likely to produce wins. It will further probe whether high or lower KDA and PKA values are likely to produce wins or not. In the game, you need to push tower down to win. On the way of win, you would kill the enemy champion or assistant your team to kill enemy, which can earn extra gold. Usually, KDA is a way to see how player plays. The higher KDA means played well or this champion have strong power at this patch Linear regression is also used to analyze the relationship between two variables, KDA and wins.

For each subject, we know both KDA (calculated) and Wins and want to find the best line of fit through the data. The resulting slope and/or intercept have a scientific meaning. A linear regression line is used as a standard curve to find new values of KDA from wins, or wins from KDAs. Linear regression and correlation are almost the same and easy to confuse. In this situation, it makes sense to do both calculations. Correlation is used to measure wins and KDA in each subject and how well they are associated.

Additionally, we calculate Pearson (parametric) correlation coefficients assuming that both wins and KDAs are sampled from normally-distributed data. It is important to note that correlation describes strength of the relationship and hence is not concerned with cause and effect.