What is the best build for Ezreal?

We do not publish a best build for Ezreal, and the reason is worth thirty seconds. The most recorded Ezreal bot build on patch 16.14 is Endless Hunger, Trinity Force and Spear of Shojin, in 481 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 Ezreal?

The most recorded Ezreal bot rune page on patch 16.14 takes Lethal Tempo out of Precision, with Presence of Mind, Legend: Haste and Cut Down beside it and Jack Of All Trades and Cash Back out of Inspiration. Players who ran it won 46.6% of 4,607 games, range 45.1% to 48.0%. 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 Ezreal?

Recorded Ezreal bot players max Q, then E, then W. 1,387 recorded games on patch 16.14.

Ezreal build, patch 16.14

5,415 games of Ezreal recorded on patch 16.14, across bot, mid, support, top and jungle.

Win rate

46.4% of bot games this patch

Pick rate

6.7% of bot games

Games

5,145 bot, patch 16.14

Rank in role

76 of 77 bot champions

Win impact against the average bot pick on patch 16.14: -2.6 wins per 100 games (range -3.8 to -1.5) 5,145 games Pooled

What most Ezreal bot players build, patch 16.14

Build Endless Hunger, Trinity Force and Spear of Shojin, and run Lethal Tempo

The rune page they run

Precision

Lethal Tempo

Inspiration

Shards

4,607 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 1,387 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 E, 11 R, 12 E, 13 E, 14 W, 15 W, 16 R, 17 W, 18 W

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

481 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 61.1% (range 56.7-65.4) · 481 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.

Ezreal bot is the role with the most recorded games on patch 16.14: 5,145 games, of which 2,385 were wins. That is the role this page leads with. Measured on this page: bot, mid, support and top, the roles that cleared 30 recorded games.

wins -2.6 per 100 vs the bot average · 5,145 games on 16.14 · first-time players measured (4,873 games) · no clear move vs 16.13 (+5.0, range -3.2 to +13.2)

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 Ezreal 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.

Ezreal, bot

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

Endless Hunger, Trinity Force and Spear of Shojin
61.1% 481 games
Trinity Force, Spear of Shojin and Serylda's Grudge
49.0% 261 games
Endless Hunger, Trinity Force, Spear of Shojin and Serylda's Grudge
60.4% 227 games
Endless Hunger, Trinity Force and Serylda's Grudge
51.9% 189 games
Trinity Force, Frozen Heart and Spear of Shojin
54.0% 50 games
Endless Hunger, Trinity Force, Frozen Heart, Spear of Shojin and Serylda's Grudge
47.6% 42 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.

Ezreal, mid

0 item sets were recorded in this role and 0 of them cleared 30 games.

No mid build recorded yet

We have 131 games of Ezreal mid on patch 16.14, but no item set repeated often enough to record as a build.

Read the rune pages instead

Ezreal, support

1 item sets were recorded in this role and 0 of them cleared 30 games.

Bastionbreaker, Umbral Glaive and Voltaic Cyclosword
not enough games yet. 8 games so far, this needs 30. 8 games

Ezreal, top

0 item sets were recorded in this role and 0 of them cleared 30 games.

No top build recorded yet

We have 48 games of Ezreal top on patch 16.14, but no item set repeated often enough to record as a build.

Read the rune pages instead

What runes do most Ezreal 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.

Ezreal, bot

10 complete pages were recorded in this role and 4 of them cleared 30 games.

Lethal Tempo Precision with Inspiration

4,607 games

Precision

Lethal Tempo

Inspiration

Shards

4,607 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

Press the Attack Precision with Inspiration

296 games

Conqueror Precision with Inspiration

108 games

Lethal Tempo Precision with Sorcery

79 games

Keystones picked, one slot on its own

Lethal Tempo: 46.6% 4,701 games
Press the Attack: 44.9% 312 games
Conqueror: 39.7% 121 games
Deathfire Touch: 56.7% 30 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.

Ezreal, mid

10 complete pages were recorded in this role and 1 of them cleared 30 games.

Lethal Tempo Precision with Inspiration

54 games

Precision

Lethal Tempo

Inspiration

Shards

54 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.

Keystones picked, one slot on its own

Lethal Tempo: 40.0% 60 games
Press the Attack: 41.9% 31 games

7 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.

Ezreal, support

8 complete pages were recorded in this role and 1 of them cleared 30 games.

First Strike Inspiration with Precision

45 games

Inspiration

First Strike

Precision

Shards

45 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.

Keystones picked, one slot on its own

First Strike: 50.9% 53 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.

Ezreal, top

5 complete pages were recorded in this role and 0 of them cleared 30 games.

Lethal Tempo Precision with Inspiration

not enough games yet. 24 games so far, this needs 30. 24 games

Precision

Lethal Tempo

Inspiration

Shards

24 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

Conqueror Precision with Inspiration

not enough games yet. 11 games so far, this needs 30. 11 games

Press the Attack Precision with Resolve

not enough games yet. 3 games so far, this needs 30. 3 games

Keystones picked, one slot on its own

Lethal Tempo: not enough games yet. 26 games so far, this needs 30. 26 games
Conqueror: not enough games yet. 12 games so far, this needs 30. 12 games
Press the Attack: not enough games yet. 6 games so far, this needs 30. 6 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 E, then W

1,387 recorded games on patch 16.14.

Ezreal bot skill path Each filled cell is the skill most players took at that level, from 1,387 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 E, 11 R, 12 E, 13 E, 14 W, 15 W, 16 R, 17 W, 18 W

Max Q, then E, then W

Only 43 recorded games on patch 16.14. A soft read. Win comparisons only count games that ran long enough to finish the max: 100% of 37 Q-first games and 100% of 4 W-first games. Short games are missing from both sides, and short games are usually decided ones.

not enough games

Ezreal mid skill path Each filled cell is the skill most players took at that level, from 43 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
E		E						E		E			E
R						R					R	R				R	R

Levels 10 to 18

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

Almost nobody maxed W first, so there is nothing to compare. +2 wins per 100 games (range -11 to 15) 41 games Measured 37 games maxed Q first, 4 games maxed W first.

Ezreal, support

Not enough support level-ups yet

not enough games yet. 11 games so far, this needs 30.

Ezreal support has 77 games recorded, but only 11 games of them carry level-ups. A levelling order off that many games is luck.

How we count level-ups

Ezreal, top

Not enough top level-ups yet

not enough games yet. 15 games so far, this needs 30.

Ezreal top has 48 games recorded, but only 15 games of them carry level-ups. A levelling order off that many games is luck.

How we count level-ups

When do players buy each piece?

The pieces of the most recorded Ezreal 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

Trinity Force: 12:30 1,005 games
Spear of Shojin: 20:00 629 games
Endless Hunger: 24:30 330 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 Ezreal?

The opponents Ezreal bot loses to most and the teammates Ezreal 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

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

Best teammates

Every measured Ezreal 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 Ezreal rank?

Every role we recorded Ezreal 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.

Ezreal by role

bot C -2.6 wins per 100 games (range -3.8 to -1.5) 5,145 games Pooled: 40 5,145 games
mid C -0.8 wins per 100 games (range -2.9 to 1.3) 131 games Pooled: 41 131 games
support C 0.0 wins per 100 games (range -2.1 to 2.2) 77 games Pooled: 45 77 games
top C +0.1 wins per 100 games (range -2.0 to 2.3) 48 games Pooled: 46 48 games
jungle not enough games yet. 14 games so far, this needs 30.: not scored 14 games

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

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.