By GamGem ResearchLast updated 20 Aug 2026 · 00:00
Zoe
the Aspect of Twilight
Mage
League of LegendsPatch 26.16MidRanked Solo/DuoEmerald+NA74,802 gamesSample data
Role snapshot (Mid)
- 53.0%
- Win Rate
- 14.6%
- Pick Rate
- 10.1%
- Ban Rate
74,802
Games
High
Confidence
Matchup map (Mid) — level 1–18
Conditional build matrix
| vs AP Burst | vs AD Assassins | vs Tanks | vs Sustain | |
|---|---|---|---|---|
| Mythic / Core |   |   |   |   |
| Core 2 |   |   |   |   |
| Core 3 |   |   |   |   |
| Boots |  |  | | |
| Situational |  |   |  |   |
Popular build
34,409 games53.4%Win Rate
Flash
IgniteReliable build
24,685 games52.1%Win Rate
Flash
TeleportHigh observed win rate
8,976 games55.5%Win Rate
Flash
ExhaustRunes
Primary: Sorcery
Arcane Comet
Axiom Arcanist
Absolute Focus
Waterwalking
Secondary: Resolve
Shield Bash
Second Wind
Shards
+9 Adaptive Force
+9 Adaptive Force
+10-180 Health (based on level)
Skill order
| Q | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| W | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
| E | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
| R | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
Ability details
More Sparkles!
Zoe's next basic attack after casting a spell deals bonus magic damage.
Passive — never cast
Paddle Star!
Zoe fires a missile that she can redirect in flight. Deals more damage the longer it flies in a straight line.
Cost 40/45/50/55/60 ManaCooldown 8.5/8/7.5/7/6.5sRange 800
Spell Thief
Zoe can pick up the remnants of enemy summoner spell and active item casts and cast them once herself. Whenever she casts a summoner spell, she gains 3 missiles that fire at the nearest target.
Cost 0 ManaCooldown 0.25sRange 3000/4500/6000/3000/3000
Sleepy Trouble Bubble
Causes the target to become drowsy, then fall asleep. While asleep, the target's Magic Resist is reduced. The first source of damage that breaks the sleep is doubled, up to a cap.
Cost 80 ManaCooldown 18/17/16/15/14sRange 800
Portal Jump
Blink to a nearby position for 1 second. Then blink back.
Cost 40 ManaCooldown 11/8/5sRange 575
Scaling — level 1–18
ZoeMid average
1918
1918
1918
Counter plan — how to beat Zoe
Dodge the primary — its rotation is one skillshot; sidestep it and the rest is chip
Punish the cooldown — Zoe is exposed while its rotation is down
Ward the flank — vision is what turns its best opening into a worse one
Respect the spike — Riot rates the kit 5/10, so a practised player gets more out of it than the average
Trade on your terms — take the fight where its damage lands on nobody
ARAM performance
- 48.5%
- Win Rate ▼ 0.1%
- 3.5%
- Pick Rate ▲ 6.2%
- 0.9%
- Ban Rate ▼ 0.6%
- Damage dealt
- 108.4%
- Damage taken
- 107.5%
- Healing
- 102.1%
- Shielding
- 107.1%
- Ability Haste
- +14
Best ARAM build
Patch timeline
We hold the rates, not the balance log. What actually changed each patch is in Riot’s own notes.
- 26.16Live now+0.8
- 26.14Sample win rate across this patch+0.0
- 26.11Sample win rate across this patch+0.3
- 26.08Sample win rate across this patch+0.4
- 26.05Sample win rate across this patch-0.2
Share cheat sheet
Methodology
- Data window
- 512,340 games on patch 26.16, Ranked Solo/Duo, NA. Last refreshed 20 Aug 2026.
- Rank aggregation
- Emerald+ and above, pooled. Splitting further leaves samples too small to separate a real effect from noise.
- Role assignment
- A champion is filed under the lanes it actually plays, not the one class it belongs to — which is why several appear in two.
- Counters
- A matchup, not a tier gap: how this champion does into that one specifically, over that pairing’s own games.
- Minimum sample
- The confidence indicator beside every rate is driven by how much of the window this champion appears in. A low reading means read the figure loosely, not that it is wrong.
- Why a build is not a cause
- The best-scoring build is the one the best-placed players happened to take. Correlation is what a rate can show; it is not evidence that buying those items caused the wins.
- Where the numbers come from
- Names, classes, lanes, abilities, items and runes are Riot Data Dragon. Win, pick and ban rates are derived samples — Riot publishes none of them, and nothing on this page presents one as measured.