infolol Editorial guide

How the infolol tier score is calculated

The formula that combines win rate and pick rate, and why a tier is not an absolute measure of champion power.

Written and reviewed by the infolol operatorPublished: Last reviewed: Editorial policy
01

Tiers compare champions only within the same role

Top samples compete with top samples, while jungle samples compete with jungle samples. Role responsibilities and pick-rate distributions differ, so a Tier 1 top champion is not directly ranked against a Tier 1 jungler.

Combinations below the 100-game publication gate are excluded from tier competition to reduce extreme small-sample results.

02

The score combines win rate and relative pick rate

The formula is (win rate - 50) × 2 + log2(pick rate ÷ average role pick rate) × 10. Win rate is centered at 50%, while pick rate is measured relative to the role's published average.

A logarithm softens the effect of extremely popular champions so that pick rate does not dominate without limit.

03

Ranked scores are divided into percentile bands

Scores are sorted and split at the top 3%, 15%, 40%, 70%, and 90% boundaries. OP additionally requires at least 52% win rate and 5% pick rate.

When only a few combinations are published, each percentile slot carries more weight. Read the coverage count on the data-status page beside the tier list.

04

What the tier does not measure

The score does not directly include ban rate, professional play value, team synergy, player mastery, or a specific matchup. A user's result can differ substantially from the aggregate rank.

This article is a general decision guide. The appropriate choice can change with the patch and team compositions.