Each option against the most picked one in these games, with the same draft odds (or game state and minute, for items), patch and player experience, shrunk as everywhere on the site. With few games, most say “Not enough games”. More filters in Explore
56% of Summon Aery games run this page. No other page wins clearly more.
Every rune for each keystone
Boots
1st item
2nd item
3rd item
Spells: Exhaust + Heal · 87% of gamesPickedHow often players take this option for this choice.Rare options often read “Not enough games” until more of their games come in.
The main number is how much more an option wins than the most picked one in similar games. Those are games with the same draft odds (for items, the same game state and minute), played on the same patch by players with as much experience. A +2.0% means about two more wins per 100 games. Numbers from few games are pulled toward what the option does on other champions.
± is the marginMargin (±)The real number is very likely inside this range. More games make it narrower.. The real difference is very likely inside it. On the bar, the dot is the number and the band is the margin. The black line means no difference.
Better or Worse means the whole margin is on one side of zero and the difference is at least half a point. Even means we know the two are within a point of each other. Unclear means it could go either way. With Not enough games, there are too few games to say anything yet.
vs expected (WPA) asks whether games with the option ended better than their win chance when it was chosen. It doesn’t compare options with each other. Next to it is the win chance it’s usually taken at.
Lane 0–14 min · Mid 14–25 · Late 25 to the end.Item numbers start at the purchase and cover the first 10 min and the rest of the game.The full guide
Players take it in 25% of games. It does clearly worse than Resolve. Most of the loss comes in lane.
Most Summon Aery games buy Ardent Censer into Mikael's Blessing or a similar support pair first (86%). Arcane Comet games lean toward Mikael's Blessing into Moonstone Renewer or a similar support pair (82%).
With support pairs like Ardent Censer into Mikael's Blessing, the data can’t tell Arcane Comet and Summon Aery games apart (−0.9% ±2.4). This describes games. It doesn’t prove a cause.
Where each rune leadsSummon Aery is the most picked. The others are compared with it. Arcane Comet is the next most picked. The Baseline row compares the two.
Summon AeryOur pickOur pickThe option shown in What wins most often: the most picked one, unless another clearly wins more.85% of gamesBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.+1.2% ±2.1vs Arcane CometLane −0.2%Mid +0.7%Late +0.7%vs expected +0.6%±2.0
Arcane Comet13% of gamesUnclear−1.2% ±2.1same players onlySame playersThe same comparison among players who used both options. Each player is compared with themselves. −15.3% ±22.5Lane +0.2%Mid −0.7%Late −0.7%vs expected Not enough gamesNot enough gamesToo few games to say anything yet. The number shows once more games come in.vs ResolveWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
ResolveBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.vs Precision+2.0% ±1.8
PrecisionWorse−2.0% ±1.8
InspirationToo close to callToo close to callPlenty of games, but too few players tried both, or the gap with the most picked option is too small to call. It could go either way.vs Ionian Boots of LucidityWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
Ionian Boots of LucidityBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.vs Ardent CenserWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
Ardent CenserBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.vs Mikael's Blessing+0.8% ±2.1
Mikael's BlessingUnclear−0.8% ±2.1
Moonstone RenewerUnclear−1.3% ±2.2
Echoes of HeliaToo close to callToo close to callPlenty of games, but too few players tried both, or the gap with the most picked option is too small to call. It could go either way.3 more options · All 1st items
vs Moonstone RenewerWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
Moonstone RenewerBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.vs Mikael's Blessing0.0% ±2.0Mikael's BlessingUnclear0.0% ±2.0Ardent CenserUnclear−0.2% ±2.0
Knight's VowUnclear+0.6% ±2.35 more options · All 2nd items
vs Moonstone RenewerWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
Moonstone RenewerBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.vs Knight's Vow0.0% ±2.4Knight's VowUnclear0.0% ±2.4Ardent CenserUnclear+1.0% ±2.4Mikael's BlessingToo close to callToo close to callPlenty of games, but too few players tried both, or the gap with the most picked option is too small to call. It could go either way.6 more options · All 3rd items
vs Knight's VowWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
Knight's VowBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.Moonstone RenewerNot enough gamesNot enough gamesToo few games to say anything yet. The number shows once more games come in.
RedemptionNot enough gamesNot enough gamesToo few games to say anything yet. The number shows once more games come in.
DawncoreNot enough gamesNot enough gamesToo few games to say anything yet. The number shows once more games come in.5 more options · All 4th items
vs Exhaust + HealWin chanceHow much more this option wins than the most picked one, in games alike in draft, patch and players. +1% is about one more win per 100 games. (most picked)
Exhaust + HealBaselineBaselineThe most picked option. The others are compared with it. Its own number compares it with the next most picked.vs Exhaust + Ignite+0.4% ±2.4
Exhaust + IgniteUnclear−0.4% ±2.4Heal + IgniteToo close to callToo close to callPlenty of games, but too few players tried both, or the gap with the most picked option is too small to call. It could go either way.
Support100%−1.7%
vs Lulu271 games+0.6% counterCounterHow the lane goes once both champions’ strength, the players and the rest of the draft are accounted for. A strong champion beating a weak one isn’t a counter.±2.4
vs Alistar85 games+0.2% counterCounterHow the lane goes once both champions’ strength, the players and the rest of the draft are accounted for. A strong champion beating a weak one isn’t a counter.±2.5
vs Bard77 games+0.2% counterCounterHow the lane goes once both champions’ strength, the players and the rest of the draft are accounted for. A strong champion beating a weak one isn’t a counter.±2.6
vs Thresh185 games−0.6% counterCounterHow the lane goes once both champions’ strength, the players and the rest of the draft are accounted for. A strong champion beating a weak one isn’t a counter.±2.5
vs Senna81 games−0.5% counterCounterHow the lane goes once both champions’ strength, the players and the rest of the draft are accounted for. A strong champion beating a weak one isn’t a counter.±2.6
vs Pantheon59 games−0.5% counterCounterHow the lane goes once both champions’ strength, the players and the rest of the draft are accounted for. A strong champion beating a weak one isn’t a counter.±2.6Best ban: ThreshFaced in 8% of games when not banned+1.3 wins per 1,000Wins savedWins a ban saves per 1,000 games: how often the enemy laner picks the champion when it’s available, times how much better they do with it.±2.2
Pool partner: AsheBetter in 23% of its lanes+0.1 wins per 1,000Wins addedWins per 1,000 games where you pick after your lane opponent, playing this champion whenever it does better against them than your main.±6.9— Not enough games to call this lane yet: its margin is wider than ±5%.
Over 2,719 ranked games on patches 16.19–16.20, Yuumi support won 47.7% and was picked in 5% of games. Most players run Summon Aery, build Ardent Censer first, Moonstone Renewer second and Knight's Vow third, and buy Ionian Boots of Lucidity. Win rates come from picks and results only. Most of the games are Emerald to Master.
Everyone or mains
Mains are players whose #1 mastery champion is this one. We know the mastery of 717,000 ranked players from Platinum to Challenger. That covers about half the players in our games. Below 1,500 games by mains, the mains view shows what they pick next to everyone’s verdicts. From 1,500 games on, every number is the mains’ own.
How the tiers are made
This mark means strength. It’s the win chance a champion adds over its role’s average once we take out who plays it.
Champions are ranked on the win chance they add over their role’s average champion once we take out who plays them. A model on every game with all ten players known accounts for each player’s mastery, rank and games on the champion, and for the rest of the draft. So a champion doesn’t look stronger because one-tricks play it, or weaker because first-timers do. Tiers: S from +2%, A from +0.75%, B from −0.75%, C from −2%, D below. The “% likely” next to a tier is how likely the champion belongs in it. Under 50% means it sits near a tier border.
WPA (win probability added)
On the pages it’s labelled “vs expected”. It compares how games with the option ended with the win chance they had when it was chosen (the draft for runes and spells, the minute of purchase for items). +1.0% is about one more win per 100 games than expected. The comparison is with what players usually do from there. It can’t compare two options with each other. Because popular options make up most of what “usual” is, it understates them. Options that strong players favor also look better here than they are. Use it as a hint and let the main number decide the verdict.
Win chance against the most picked
This is how much more each option wins than the most picked one, in points of win chance. The games compared have the same draft odds or game state, the same patch and the same player experience. Numbers from few games are then pulled toward what the same option does on other champions. A number from 300 games keeps little of its size. With thousands of games, most of it stays.
Better, worse, even, unclear
The verdict reads the number after the pull and its 95% margin (±). Better or Worse needs the whole margin on one side of zero and a difference of at least half a point. Even means we know the two are within a point of each other. When it could go either way, the verdict is Unclear. Not enough games means too few players tried both, or the number is mostly the pull.
Lane, mid and late game
The difference is split by when the win chance moved: lane (0–14 min), mid game (14–25) and late game (25 to the end). The three add up to the whole game. They show when an option pays off or costs you.
Our pick
We pick an option over the most picked one only if it’s in 5% or more of games and reads Better. It must also stay ahead when checked on a separate half of the games. Otherwise we pick the most picked option. When a second option reads Even with it (within a point), we show both.
Not enough games
There are too few games to say anything yet. The margin would be wider than ±5%, or the number would be too noisy to say anything. The number appears once more games narrow the margin.
What the data says
We run seven checks on every champion. Two flag common picks that cost win chance and results carried by one player. The others catch theory that real games don’t back, answers that change with your build, keystone gaps that go with the build, options that pay off in only one phase, and lanes that turn after minute 14.
Where each rune leads
A build here is the kind of pair your first two legendary items make, read from the items’ stats. Bruiser means both give health, armor or magic resist and at least one gives damage. Burst means both give ability power and nothing that keeps you alive. Mixed is one of each. Tank, crit, lethality, damage and support pairs follow the same rule. Each build is named by its most common pair. The shares are plain counts of games. Items are bought after the rune is chosen. Two keystones compared inside one build describe those games. They don’t prove the rune caused the result.
Counter
A counter is how this lane goes after the model accounts for both champions’ overall strength, who played them (mastery, rank, games on the champion) and the rest of the draft. Lanes with few games are pulled toward zero. A strong champion beating a weak one doesn’t count as a counter. The counter uses only games where all ten players are known. It rests on fewer games than the win rate next to it.
Against the usual win rate
This is the champion’s win rate against this opponent minus its win rate over all its games in the role. It uses picks and results only. Splitting lane from late game needs games with full timelines.
Common but costly
Players take this option in at least 10% of games. It does clearly worse than the most picked option. Even after the pull, its whole margin sits below zero.
