search

jeffseo / ChampionMasteryPlanner

A tool for League of Legends users to plan out how many games are needed to obtain the champion points they want
1 stars 0 forks source link

Champion point estimation formula #7

Closed jeffseo closed 8 years ago

jeffseo commented 8 years ago

Come up with formula to calculate how much points is expected for win and loss.

g5qinyix commented 8 years ago

We can come up with a formula and give three possible estimations to the users: 1) Minimum number of wins required (If user wins 100%) 2) Maximum number of wins required (If user loses 100%) 3) How many games required assuming 50% win/loss rate

g5qinyix commented 8 years ago

Another issue related to this comes up. Our current model/formula only takes into account LENGTH OF GAME and WIN/LOSS. It seems like we want to give the user an estimation of THE NUMBER OF GAMES NEEDED TO REACH MASTERY LEVEL ____. Therefore, we need to estimate the length of each game the user needs to play in the future. Is 33-35 minutes a good estimate? I took average length of a game for the data we have and it comes out to be ~34 minutes.

g5qinyix commented 8 years ago

(The top 10 page's games needed column can be the assumed 50% win rate, just for simplicity. If the user wants to see games needed based on other winrates, they would need to go to the individual champion page).

Another idea for the individual champion page: Rather than assuming the three scenarios (100% win rate, 50% win rate, 0% win rate) on the individual champion page, we can provide a slide bar for the user. This slide bar represents a theoretical win rate and varies from 0 to 100 with increments of 10. I.e. If the user wants to see how many games they need to play assuming they win 50%, they would move the slider to 50% mark, our formula is used and a number (representing how many games are required to be played) is given to the user. The user can then slide the bar down to like 40% (or up to 60%) and the number will be updated.

Alternatively if slider too hard to implement (?): We allow the user to input their own theoretical win rate as a float and we just provide the number of games needed based on the user inputted winrate.

g5qinyix commented 8 years ago

Closing this as formula itself has been determined.