LoL Tier ListPatch 26.22

Patch 26.22Updated 14 Nov 2026

How the tier list is made

Where the numbers come from, how they are combined into one score and tier, how far the big stat sites can be trusted, and what a tier list cannot tell you.

Same numbers everywhere, very different grades

At the same rank every site reads the same Riot match data, so their win rates nearly match. Their letter grades do not. Below is how closely each site's tiers follow winning versus popularity, as rank correlations across about 200 champion-role pairs at Diamond+.

Last column: how closely the site's win rates match OP.GG at the same rank (1 = identical). Mobalytics' expert list has no win rates.

U.GG and Blitz are the sites whose letters you can take at face value. dpm.lol, Mobalytics' expert list and METAsrc mostly tell you what people pick. That is why Kai'Sa, Lee Sin, Lulu and Nautilus sit in their top tiers while winning less than the average pick in their role.

How the score works

  • Per dataset: win rate minus that dataset's average for the role, in percentage points.
  • Sample size: each number is pulled toward zero by how noisy it is. Game counts come from OP.GG's sample totals and U.GG. How hard to pull is fitted from the data for every dataset and role, not guessed.
  • One-tricks: a small penalty under 2% pick rate, because mostly mains play those champions and their win rate flatters the champion.
  • Overall view: Global 30%, EUW 20%, KR 20%, Master+ 20%, NA 10% (Emerald+ unless stated).
  • High elo: Diamond+ from OP.GG, U.GG, Blitz, dpm.lol and League of Graphs at half weight (40%), Master+ from OP.GG and U.GG (35%), EUW Diamond+ and Master+ (25%).
  • Tiers follow fixed cutoffs, no hand edits: S from +1.15 and above average in every region, A from +0.6, B from +0.15, C from −0.45. Z needs +1.8 overall, +1.5 or better in every dataset and a 2%+ pick rate. Under 1.5% pick and strong goes to P.
  • Last patch is the starting point for champions the new patch did not touch, so early-patch numbers don't jump around. How far a champion can move between patches is fitted from the data (about 0.2 points).
  • The ± next to a score is a rough 90% range. Arrows show the change in Global Emerald+ since last patch.

What it cannot tell you

  • Your champion pool. A one-trick on a B-tier pick beats a first-timer on a Z.
  • Matchups and draft. Mage bots and Kog'Maw need the right support.
  • Chinese servers. Tencent's data lives in its apps and is not public.
  • Anything after 26.23. Master Yi, LeBlanc and Syndra get nerfed on 18 Nov.
  • Game counts for Blitz and League of Graphs. They don't show them, so OP.GG Diamond+ counts stand in.
  • How good the players are on the champion. That would need per-player data from the Riot API.

Sites checked

SiteHomeWhat I readVerdict

Patch info