LoL Matchup Simulator
AI-Powered Match Prediction
Enter two Riot IDs and let our algorithm analyze real stats to predict who would win.
---
---
How It Works
Enter Riot IDs
Type the Riot IDs of two players (or two teams of five) and select the server they play on.
Data Analysis
Our system fetches real-time data from Riot API including rank, recent match history, KDA, champion mastery, and more.
Score Calculation
A multi-factor scoring algorithm calculates each player's Core Score based on tier, win rate, KDA, farm efficiency, champion mastery, and consistency.
AI Prediction
An Elo-based formula converts score differences into win probabilities. Win/loss is determined purely by scores; AI generates commentary only.
Scoring System Explained
Our Core Score algorithm evaluates players across six key dimensions. Each component is normalized to a 0-1000 scale and weighted to produce a final score. A tier calibration factor is applied to prevent overestimation of lower-tier players.
Tier
Ranked tier and LP are converted to a numerical score. Iron starts at 100, Challenger at 1000+. A calibration coefficient adjusts for opponent-relative performance, preventing low-tier inflation.
Win Rate
Recent 20-game win rate multiplied by 1000. A 65% win rate equals 650 points.
KDA
Average KDA normalized against 5.0. A KDA of 3.5 yields 700 points.
Farm
CS per minute and gold per minute combined. 8 CS/min + 500 gold/min = max score.
Mastery
Top 3 champion mastery points averaged. 500k average mastery = max score.
Consistency
Lower performance variance means higher consistency bonus.
Core Score = (Tier x 0.30) + (WinRate x 0.25) + (KDA x 0.20) + (Farm x 0.10) + (Mastery x 0.10) + (Consistency x 0.05) Frequently Asked Questions
How does the matchup scoring work?
We use a multi-factor scoring algorithm that weighs six key stats: ranked tier (30%), recent win rate (25%), KDA (20%), farming efficiency (10%), champion mastery (10%), and performance consistency (5%). The scores are then converted to win probabilities using an Elo-based formula.
What data is used to predict the winner?
We pull real-time data from Riot's official API: ranked information, recent 20 match history, champion mastery stats, and general account data. No third-party or unofficial data is used.
Is this an official Riot Games tool?
No. This is a fan-made simulation tool for entertainment purposes only. It is not endorsed by Riot Games.
How accurate are the predictions?
Predictions reflect statistical tendencies based on recent performance. Many in-game variables (champion picks, team composition, mental state) are not captured, so treat results as educated estimates rather than guarantees.
What is the difference between 1v1 and 5v5 mode?
In 1v1 mode, two individual players are compared directly. In 5v5 mode, five players per team are assigned positions (Top, Jungle, Mid, ADC, Support) and the algorithm compares each lane plus overall team strength.
Which servers/regions are supported?
All 17 Riot servers are supported: KR, NA1, EUW1, EUN1, JP1, BR1, LA1, LA2, TR1, RU, ME1, OC1, PH2, SG2, TH2, TW2, VN2. Routing: ASIA (KR, JP1), AMERICAS (NA1, BR1, LA1, LA2), EUROPE (EUW1, EUN1, TR1, RU, ME1), SEA (OC1, PH2, SG2, TH2, TW2, VN2).
How does the tier calibration work?
A calibration coefficient adjusts scores to prevent low-tier overestimation. Iron/Bronze receive a 0.9x factor, Silver 0.95x, and Gold+ is 1.0x. This corrects for the statistical tendency of lower-ranked players to appear stronger than they are relative to higher-ranked opponents.
Does AI decide the winner?
No. Win/loss is determined purely by the score-based algorithm (Elo formula). AI only generates the match commentary/narrative after the result is calculated. The prediction is 100% data-driven.
LoL Matchup Simulator is not endorsed by Riot Games and does not reflect the views or opinions of Riot Games or anyone officially involved in producing or managing Riot Games properties. Riot Games and all associated properties are trademarks or registered trademarks of Riot Games, Inc.