By GamGem ResearchLast updated 20 Aug 2026 · 00:00

Twitch
the Plague Rat
MarksmanAssassin
League of LegendsPatch 26.16BotRanked Solo/DuoEmerald+NA11,784 gamesSample data
Role snapshot (Bot)
- 46.5%
- Win Rate
- 2.3%
- Pick Rate
- 22.3%
- Ban Rate
11,784
Games
Moderate
Confidence
Matchup map (Bot) — 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
5,421 games46.9%Win Rate
Reliable build
3,889 games45.6%Win Rate
High observed win rate
1,414 games49.0%Win Rate
Runes
Primary: Precision
Press the Attack
Presence of Mind
Legend: Bloodline
Coup de Grace
Secondary: Sorcery
Axiom Arcanist
Absolute Focus
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

Deadly Venom
Twitch's basic attacks infect the target %i:OnHit% On-Hit, dealing true damage each second.
Passive — never cast

Ambush
Twitch becomes Camouflaged for a short duration and gains Move Speed. When leaving Camouflage, Twitch gains Attack Speed for a short duration.When an enemy champion with Deadly Venom dies, Ambush's cooldown is reset.
Cost 40 ManaCooldown 16sRange 20

Venom Cask
Twitch hurls a cask of venom that explodes in an area, slowing targets and applying deadly venom to the target.
Cost 70 ManaCooldown 13/12/11/10/9sRange 950

Contaminate
Twitch wreaks further havoc on poisoned enemies with a blast of his vile diseases.
Cost 50/60/70/80/90 ManaCooldown 12/11/10/9/8sRange 1200

Spray and Pray
Twitch unleashes the full power of his crossbow, shooting bolts over a great distance that pierce all enemies caught in their path.
Cost 100 ManaCooldown 90sRange 1200
Scaling — level 1–18
TwitchBot average
1918
1918
1918
Counter plan — how to beat Twitch
Close the gap early — it wins at range and at three items — neither before ten minutes
Deny the reset — it wants a kill to come back out; a warded exit removes it
Punish the cooldown — Twitch 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 6/10, so a practised player gets more out of it than the average
ARAM performance
- 49.8%
- Win Rate ▲ 0.6%
- 2.8%
- Pick Rate ▼ 2.1%
- 0.6%
- Ban Rate ▼ 1.5%
- Damage dealt
- 109.8%
- Damage taken
- 106.5%
- Healing
- 104.5%
- Shielding
- 91.5%
- Ability Haste
- +7
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-0.7
- 26.14Sample win rate across this patch-0.8
- 26.11Sample win rate across this patch-0.8
- 26.08Sample win rate across this patch-0.2
- 26.05Sample win rate across this patch+0.9
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.
