What is LoL Recommender?
- LoL Recommender is League of Legends champion recommendation service that helps you find new champions to play.
- Get recommended similar champions using your summoner name, a champion you like, or filter champions by a tag or combination of several tags.
- Developed using the Orianna Java Framework
- Relevant Champion Data
- Recommended champions are only champions that are currently being played in Plat+ ranked games.
- Every champion recommendation shows the highest win rate Summoner Spells, Builds, Skill Order, and Masteries for that champion based on Plat+ ranked games in the current patch (11.8).
- Recommended champions will consist of different lanes/positions but you can use the filter to only show recommendations for a particular lane.
- If you use this filter, the recommended champions will have the highest win rate build corresponding to the filtered position.
- When using a champion for recommendations, the highest win rate info for the current patch (11.8) is displayed at the top of the page for the selected champion.
How does it work?
- The recommendation algorithm dynamically assigns weighted tags to each champion based on the information available in the Riot API.
- Weighted Tags
- The tags are mapped to the API every time you load a page. This ensures that LoL Recommender stays up to date.
- The weighted tags are mapped to a champion using a combination of text analysis and manual filters.
- Champion Recommendations
- When recommending by champion, the recommendation algorithm compares the selected champion with all other League of Legends champions that are currently being played in Plat+ ranked games.
- Each champion is assigned a match score with all other champions.
- The 8 champions with the highest match score are recommended (must be >50% match score).
- Summoner Name Recommendations
- When recommending by Summoner name, the recommendation algorithm grabs the Summoner's top 2 most played champions of all time (using their mastery score) and the most played champions from their last 100 matches (max of 6 champions).
- The champion recommendation algorithm is then run on each of the 8 champions and the resulting recommended champions are grouped together, sorted by match scores, and then the 8 champions with the highest match score are recommended.
- Match Score Calculation
- The Match Score is calculated using a combination of the number of tags in common and the total weight of all the shared tags.
- Donate - I am solo developing and hosting LoL Recommender off of my laptop so please consider donating so I can keep it going and add to it!
- Contact - If you notice anything wrong with the site, have feedback, want to advertise, or would like your art featured, please contact me.
- Art Source - All of the art is currently coming from the League of Legends Wiki.