What is the best build for Xerath?

We do not publish a best build for Xerath, and the reason is worth thirty seconds. The most recorded Xerath bot build on patch 16.14 is Horizon Focus, Shadowflame and Luden's Echo, in 323 games. A finished item set is read at the end of the game, so a build only appears once the game ran long enough to buy it, and a game that runs long is usually a game you were winning. Any share of wins taken from those games leans high for that reason on its own, which is why there is not one in this answer. The page lists every recorded build with the games behind it.

What are the best runes for Xerath?

The most recorded Xerath bot rune page on patch 16.14 takes Arcane Comet out of Sorcery, with Manaflow Band, Transcendence and Scorch beside it and Triple Tonic and Cash Back out of Inspiration. Players who ran it won 51.9% of 1,495 games, range 49.4% to 54.4%. A rune page is locked in at champion select, so that share is not inflated by long games the way a build is, but it counts who picked the page rather than what the page did.

What skill order should I use on Xerath?

Recorded Xerath bot players max Q, then W, then E. 570 recorded games on patch 16.14.

Xerath build, patch 16.14

4,489 games of Xerath recorded on patch 16.14, across bot, mid, support and top.

Win rate

52.5% of bot games this patch

Pick rate

3.3% of bot games

Games

2,545 bot, patch 16.14

Rank in role

1 of 77 bot champions

Win impact against the average bot pick on patch 16.14: +1.4 wins per 100 games (range -0.1 to 2.8) 2,545 games Pooled

What most Xerath bot players build, patch 16.14

Build Horizon Focus, Shadowflame and Luden's Echo, and run Arcane Comet

The rune page they run

Sorcery

Arcane Comet

Inspiration

Shards

1,495 games ran this page exactly as it is drawn. We never stitch the most picked keystone onto the most picked second row and call it a page: nobody ran that one.

The skill path Each filled cell is the skill most players took at that level, from 570 games.
skill	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
Q	Q			Q	Q		Q		Q
W		W						W		W		W	W
E			E											E	E			E
R						R					R					R	R

Levels 10 to 18

10 W, 11 R, 12 W, 13 W, 14 E, 15 E, 16 R, 17 R, 18 E

Max Q, then W, then E. The level by level count.

323 games finished this exact set. It is the set recorded most often, which is a fact about players rather than about the items.

Share of those games that were wins, in percent 50.5% (range 45.0-55.9) · 323 games · Measured

Games that finish a build are games that lasted, and lasting usually means winning, so this figure leans high before the items do anything. It is how those games went, not what the build does to your chances. How far that can bend a number.

Xerath bot is the role with the most recorded games on patch 16.14: 2,545 games, of which 1,336 were wins. That is the role this page leads with. Measured on this page: bot, mid and support, the roles that cleared 30 recorded games.

wins +1.4 per 100 vs the bot average · 2,545 games on 16.14 · first-time players measured (2,363 games) · no clear move vs 16.13 (+2.0, range -9.4 to +13.4)

The build above is the one recorded most often, nothing more. Every recorded build, with its own games and its own share, is in the build list, in the order players built them.

Why this answer

A champion and role pair needs 30 recorded games before we put a score or a win impact on it: under that, a handful of lucky games moves a win rate by ten points. The same floor applies to a single build and to a single rune page, and anything under it prints its games and no number.

The share on that figure says how the games went, not what the build does to you. Nothing on this page measures that; the method note at the end says why.

What do most Xerath players build?

Item sets recorded on patch 16.14, most built first, never sorted by how the games went. The share is how many of a build's own games were wins; a build under 30 games shows its games and no share.

Xerath, bot

8 item sets were recorded in this role and 8 of them cleared 30 games.

Horizon Focus, Shadowflame and Luden's Echo
50.5% 323 games
Rabadon's Deathcap, Horizon Focus, Shadowflame and Luden's Echo
57.9% 152 games
Rabadon's Deathcap, Shadowflame and Luden's Echo
53.2% 124 games
Rabadon's Deathcap, Horizon Focus and Luden's Echo
60.9% 64 games
Rabadon's Deathcap, Void Staff, Horizon Focus, Shadowflame and Luden's Echo
50.0% 58 games
Zhonya's Hourglass, Horizon Focus, Shadowflame and Luden's Echo
51.0% 49 games

A finished set is read at the end of the game, so every share here leans high: games that last long enough to finish a build are usually games that were already going well. How far that can bend a number.

Xerath, mid

8 item sets were recorded in this role and 6 of them cleared 30 games.

Rabadon's Deathcap, Shadowflame and Luden's Echo
47.1% 153 games
Horizon Focus, Shadowflame and Luden's Echo
54.9% 91 games
Shadowflame, Stormsurge and Luden's Echo
67.6% 34 games
Rabadon's Deathcap, Zhonya's Hourglass, Shadowflame and Luden's Echo
65.6% 32 games
Rabadon's Deathcap, Void Staff, Shadowflame and Luden's Echo
35.5% 31 games
Rabadon's Deathcap, Horizon Focus, Shadowflame and Luden's Echo
50.0% 30 games

Xerath, support

6 item sets were recorded in this role and 1 of them cleared 30 games.

Horizon Focus, Shadowflame and Luden's Echo
62.9% 35 games

What runes do most Xerath players run?

Whole pages that players actually ran, most recorded first. A page is locked in at champion select, so long games do not inflate its share the way they inflate a build's; it still counts who picked the page rather than what the page did.

Xerath, bot

10 complete pages were recorded in this role and 3 of them cleared 30 games.

Arcane Comet Sorcery with Inspiration

1,495 games

Sorcery

Arcane Comet

Inspiration

Shards

1,495 games ran this page exactly as it is drawn. We never stitch the most picked keystone onto the most picked second row and call it a page: nobody ran that one.

Other pages players ran

Arcane Comet Sorcery with Precision

856 games

First Strike Inspiration with Sorcery

139 games

Keystones picked, one slot on its own

Arcane Comet: 52.1% 2,383 games
First Strike: 54.4% 147 games

5 keystones were recorded in this role: one slot on its own, counted across every page that used it, so the share carries the rest of each page with it.

Xerath, mid

10 complete pages were recorded in this role and 5 of them cleared 30 games.

Arcane Comet Sorcery with Precision

751 games

Sorcery

Arcane Comet

Precision

Shards

751 games ran this page exactly as it is drawn. We never stitch the most picked keystone onto the most picked second row and call it a page: nobody ran that one.

Other pages players ran

Arcane Comet Sorcery with Inspiration

220 games

First Strike Inspiration with Sorcery

215 games

Arcane Comet Sorcery with Domination

60 games

Dark Harvest Domination with Sorcery

44 games

Keystones picked, one slot on its own

Arcane Comet: 47.7% 1,046 games
First Strike: 53.1% 224 games
Dark Harvest: 46.0% 50 games

5 keystones were recorded in this role: one slot on its own, counted across every page that used it, so the share carries the rest of each page with it.

Xerath, support

8 complete pages were recorded in this role and 3 of them cleared 30 games.

Arcane Comet Sorcery with Precision

304 games

Sorcery

Arcane Comet

Precision

Shards

304 games ran this page exactly as it is drawn. We never stitch the most picked keystone onto the most picked second row and call it a page: nobody ran that one.

Other pages players ran

First Strike Inspiration with Sorcery

131 games

Arcane Comet Sorcery with Inspiration

112 games

Keystones picked, one slot on its own

Arcane Comet: 44.7% 436 games
First Strike: 43.0% 135 games
Dark Harvest: 45.5% 33 games

4 keystones were recorded in this role: one slot on its own, counted across every page that used it, so the share carries the rest of each page with it.

What skill order should I use?

Counted from recorded level-ups on patch 16.14, one player per game.

Max Q, then W, then E

570 recorded games on patch 16.14.

Xerath bot skill path Each filled cell is the skill most players took at that level, from 570 games.
skill	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
Q	Q			Q	Q		Q		Q
W		W						W		W		W	W
E			E											E	E			E
R						R					R					R	R

Levels 10 to 18

10 W, 11 R, 12 W, 13 W, 14 E, 15 E, 16 R, 17 R, 18 E

Max Q, then W, then E

318 recorded games on patch 16.14.

Xerath mid skill path Each filled cell is the skill most players took at that level, from 318 games.
skill	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
Q	Q			Q	Q		Q		Q
W		W						W		W		W	W
E			E											E	E		E	E
R						R					R					R

Levels 10 to 18

10 W, 11 R, 12 W, 13 W, 14 E, 15 E, 16 R, 17 E, 18 E

Max Q, then W, then E

153 recorded games on patch 16.14. Win comparisons only count games that ran long enough to finish the max: 99% of 137 Q-first games and 88% of 16 W-first games. Short games are missing from both sides, and short games are usually decided ones.

not enough games

Xerath support skill path Each filled cell is the skill most players took at that level, from 153 games.
skill	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
Q		Q		Q	Q		Q		Q
W	W							W		W		W	W
E			E											E	E		E	E
R						R					R					R

Levels 10 to 18

10 W, 11 R, 12 W, 13 W, 14 E, 15 E, 16 R, 17 E, 18 E

Almost nobody maxed W first, so there is nothing to compare. 0 wins per 100 games (range -11 to 12) 150 games Measured 136 games maxed Q first, 14 games maxed W first.

When do players buy each piece?

The pieces of the most recorded Xerath bot build, in the order players buy them. The time is the middle of the spread across every recorded game on patch 16.14, measured at the buy rather than at the end of the game: half the recorded buys of the item happen before it, half after.

Buy times, earliest first

Luden's Echo: 11:06 1,124 games
Horizon Focus: 19:47 202 games
Shadowflame: 20:43 1,918 games

The bar is scaled to the latest piece of the set. Each item links to its own page, where the whole spread sits beside the games behind it.

Who counters Xerath?

The opponents Xerath bot loses to most and the teammates Xerath gains the most beside on patch 16.14. The chip is the edge in wins per 100 games, and the games behind it sit under the name.

Hardest opponents

71 more opponents sit under 50 games against Xerath and are counted, not ranked. Every measured Xerath opponent, with the full table

Best teammates

Every measured Xerath teammate, with the full table

These numbers come from 108,616 ranked games we recorded on EUW, Korea and NA, found by following Diamond players and the people they play with. That is not every rank and it is not every region. Everything on this page is patch 16.14. Counted up to 24 July 2026.

How does Xerath rank?

Every role we recorded Xerath in on patch 16.14, scored or not. Win impact is measured against the average pick in that role on the same patch, and a role needs 30 recorded games before it is scored.

Xerath by role

bot A +1.4 wins per 100 games (range -0.1 to 2.8) 2,545 games Pooled: 64 2,545 games
mid C -0.5 wins per 100 games (range -2.2 to 1.2) 1,332 games Pooled: 46 1,332 games
support C -1.3 wins per 100 games (range -3.2 to 0.6) 605 games Pooled: 40 605 games
top not enough games yet. 7 games so far, this needs 30.: not scored 7 games

Why no share of wins on this page says a build is good for Xerath

A finished item set is read at the end of the game. A player who bought the first two items and lost at twenty minutes never finished the set, so they are not in its row. Everybody in the row lasted long enough to buy it, and lasting is most of what winning looks like. The share those games won is pushed up by that on its own, before the items do anything at all, and the push is largest on the most expensive builds and backwards on the ones you buy when a game is going badly.

We know the size of it. In simulation, with every item's real effect set to exactly zero, an item read at the end of the game shows about 92% as a sixth completed item. A defensive item with a real 5 point benefit shows about 20 points the wrong way, because it gets bought when a player is behind. The measurement is here.

So this page counts and labels rather than ranks. The estimate that answers the real question is taken at the moment of the buy, from the state the purchase happened in, and it runs in the desktop app because it needs the game in front of it.