Urgot has the highest win rate out of all ADCs this patch

Bârd·May 28, 2016·58 votes·4,773 views

http://www.grimmemennesker.dk/data/media/2/wat.jpg

40 Comments

WoonStruckMay 28, 201639 votes

17 people getting lucky and winning.

Best Kog NAMay 28, 201638 votes

{quoted}

7 people getting lucky and winning.

I fixed it

EndlessSorcererMay 28, 20167 votes

I think you may have removed the wrong digit.

Troll for TrumpMay 28, 201610 votes

1 person getting lucky and winning.

There.

Bobloser3555May 28, 20165 votes

Can't be if it was just one person it would be 100% win rate.

Count ZubatMay 28, 20168 votes

Didn't that happen once with Yorick in high elo statistics? I'm pretty sure it did.

Dan AbnormalMay 28, 20162 votes

it happened to yorick urgot and karthus in challenger

DefendedMay 28, 20167 votes

a .1% playrate doesnt account for any win rate.

ZarvanisMay 28, 20165 votes

Holy shit, he has a 54% win rate in Platinum+ right now. What they hell happened that made him so powerful?

Lady HammerlockMay 28, 201625 votes

{quoted}

Holy shit, he has a 54% win rate in Platinum+ right now. What they hell happened that made him so powerful?

Urgot is the exact same as Karthus Whatever left of his player base are the true main mains which means they're really good at him

Avios1May 28, 20161 votes

I used to be an Urgot main, now he is just my off-main.

I main Teemo now, might wanna switch back to Urgot.

kile147May 28, 20165 votes

ok Satan

Sire HippingtonMay 28, 20165 votes

wiered thing is that most urgot adc players are new to him, over 50% of the players have less than 5 games on him http://champion.gg/champion/Urgot/ADC

Karthus has 30% new players http://champion.gg/champion/Karthus/Top

For lucian it's only ~8% of the players http://champion.gg/champion/Lucian

It seems logical that low pickrate means only experienced players play the champ, but more often than not this isn't the case

The main reason for their winrates is the extreme small samplesize of games which makes the statistically irelevant. For comparison, role a dice 24 times and note the results. Theoretically there is an equal distribution between the numbers, so 1-6 each 4 times, but in practice that usually isn't the case due to small sample sizes

Hyrum GraffMay 28, 201617 votes

[{quoted}](name=Don Pavoni,realm=EUW,application-id=3ErqAdtq,discussion-id=fXqsq2Xs,comment-id=000100000001,timestamp=2016-05-28T11:16:22.507+0000)

wiered thing is that most urgot adc players are new to him, over 50% of the players have less than 5 games on him http://champion.gg/champion/Urgot/ADC

That's actually exactly what you would expect, because that's a graph of players and not a graph of games played.

With some huge rounding: 50% of Urgot players have 1-5 games. Let's assume 3 is the average.

7.5% have 50-125 games. Let's assume 60 is the average.

50% x 3 = 150gpi (games played index) 7.5% x 3 = 450gpi

Even though most people playing Urgot are new to him, there are 3 times more games being played by the 50-125 category than the 1-5 category.

What the graph actually shows is something called churn. See ZenonTheStoic's Taliyah Q&A for more about that. Churn measures the answer to "Do players who try Urgot stick with him?"

The answer is a resounding NO. Urgot has a reaaaaly high rate of churn: 50% of players try him out and then never play him again.

Compare that to Lucian's graph. Most people who try lucian, play him more than 5 times. The 15-50 category (assume average 30) is 35%. 30 x 35% = 1050gpi

Compare that to the 150gpi we got above for Urgot above, or with 1500gpi for Lucian's 50-125 (made up avg 60) category. A much higher % of games played on lucian are players with less than 50 games of experience, compared to Urgot players, though this is somewhat offset by there also being a higher % of lucian mains.

If I'm still interested later when I'm on a computer I'll make a few graphs to visualize this.

Sire HippingtonMay 28, 20161 votes

Fair enough, though it would be quite interesting to have data for the exact distribution of games played for the different 'experince levels', since your numbers seem a bit off as the overall winrate would be to small for that.

Let's use your average numbers and add some for the other experience levels,
the 50% ~3 game average have a ~50% winrate, 150gpi(games played index, (player-%) x (average number of games played) ) the 15% ~10 games average have ~61%, 150 gpi the 15% ~30 games average have ~66%, 450 gpi the 10% ~60 games average have ~70%, 600 gpi the 10% ~150 ganes average have ~80% 1500 gpi based on the champion.gg graph. the player% numbers differ a bit to both grant easier numbers and because the champion.gg numbers don't add up to 100% somehow

total gpi: 2850

so the 150 games average make up for ~53% of all games and thus the total winrate the 60 games average make up for ~21%, the 30 games average make up for ~15% the 10 and 3 games average each make up for ~5%

which means the total winrate would be (53 x 80 + 21 x 70 + 15 x 66 + 5 x 61 + 5 x 50)/100 = ~72.55 Even if we take 10% because of roughly estimated numbers, it would still be far higher than the overall winrate of ~54% stated by champion.gg, and since the average games per player total with my numbers woukld be only 28.5 rather than 53.9, it is rather save to asume that the estimated values for games average are far to low actually, and higher average games would lead to a higher winrate, so in the end, your aproch most likely fails at determinating the %games played of each experience level

The games played stat is collected over a longer time frame than one patch, which is why your aproach doesn't work out.

That also makes sense, because a player that has 50+ urgot games total this patch will also have 50+ games in the next patch and so on, the long term playetrs stay the same each patch with only a few maybe leaving and a few new mains comming, but the 1-5games players that just try him out change each patch, as they most likely do their 1-5 games in close succesion and then either abandon urgot or stay with him.

Still, i think the drastic swings in winrate without big changes(neither urgot nor items he uses were changed that patch) are more likely a statistic anomally caused by small sample size, else his winrate would have changed in all roles, or it would have been on that level befor that patch already, since i don't think that the ratio of games form experienced urgots to games played by inexperienced urgots has really changed that much within that one patch, especially not towards that direction.

Hyrum GraffMay 28, 20161 votes

[{quoted}](name=Don Pavoni,realm=EUW,application-id=3ErqAdtq,discussion-id=fXqsq2Xs,comment-id=00010000000100000000,timestamp=2016-05-28T19:40:45.353+0000)

gpi

Calling it this instead of a % is so much nicer, good call. Edited my post above to use gpi.

the player% numbers differ a bit to both grant easier numbers and because the champion.gg numbers don't add up to 100% somehow_

They forgot to count the players with 0 games played kappa. (but actually, that's 2014 & 2015 ranked; what about the players who just picked up urgot this season?)

so the 150 games average make up for ~53% of all games and thus the total winrate the 60 games average make up for ~21%, the 30 games average make up for ~15% the 10 and 3 games average each make up for ~5% Even if we take 10% because of roughly estimated numbers, it would still be far higher than the overall winrate of ~54% stated by champion.gg, and since the average games per player total with my numbers would be only 28.5 rather than 53.9, it is rather save to asume that the estimated values for games average are far to low actually, and higher average games would lead to a higher winrate, so in the end, your aproach most likely fails at determinating the %games played of each experience level

Where are you getting the 28.5 average games per player statistic?

The games played stat is collected over a longer time frame than one patch, which is why your aproach doesn't work out.

I'm pretty sure the point of "Most people playing urgot have played very little urgot, therefore experienced urgot players hold a high % of the total gpi" still holds. However, you're definitely right that those specific numbers won't work, because the pie chart only considers who's playing urgot this patch, not how many games each of those people are playing on urgot this patch.

Sire HippingtonMay 28, 20161 votes

Where are you getting the 28.5 average games per player statistic?

50% of the players have 3 games, 15% have 10, 15% have 30, 10% have 60 and 10% have 150 by my estimated numbers. to get the overall average games, you multiply the % with the games(which also is the gpi), ad it all together and then divide it by the total %(100) so (50 x 3 + 15 x 10 + 15 x 30 + 10 x 60 + 10 x 150)/100 = 2850/100 = 28.5

Dr PoroMay 29, 20162 votes

I thought this stat was calculated as: Each time you play a game, you will be another "player" for this statistic. So someone who played urgot 100 times already and plays him 10 times in this patch will count as 10 players for urgot with 100-110 games played. Because that's the experience he's had with urgot in each of those games.

Concrete for urgot: In 52% of the games played, the player playing Urgot has played 0-4 Urgot games before (since season 2014). Only 8% of the urgot games this patch have been played by 'Urgot mains' with 125+ ranked urgot games in their history.

If it was really the playerbase average, most champs would have a way lower score because the majority of the playerbase hasn't even played a certain champion.

Sire HippingtonMay 29, 20161 votes

Each time you play a game, you will be another "player" for this statistic. So someone who played urgot 100 times already and plays him 10 times in this patch will count as 10 players for urgot with 100-110 games played. Because that's the experience he's had with urgot in each of those games.

This seems to be the most likely explanation so far.

charcharmunroMay 28, 20163 votes

The fact that less than 400 games were played with him as ADC, compared to the 800-ish games played as Top where he's more "generally effective". Also... I guess the glitch on his ult that made the slow or fear last an insane time might have been a factor. Azir's win-rate skyrocketed when he had a glitch that let his soldiers do insane damage to a single target.

CurlyFriezzzMay 29, 20161 votes

{quoted}

Holy shit, he has a 54% win rate in Platinum+ right now. What they hell happened that made him so powerful?

Probably Black Cleaver price reduction.

The DeckownerMay 28, 20163 votes

remember he had a bug this patch where he perma slow/fear people.

The LexerMay 28, 20161 votes

Shhh. Let Urgot get his glory

Twitch ChatMay 28, 20163 votes

I'm going to make the assumption that the champion.gg stats are counting the 30s fear bug

kaltenashesMay 28, 20166 votes

wasnt it a 365 days fear bug?

Hyrum GraffMay 28, 201610 votes

{quoted}

wasnt it a 365 days fear bug?

365.25

Voluug13May 28, 20162 votes

358/2 days of fear.

Sire HippingtonMay 28, 20161 votes

if it was caused by that bug, one would think urgot in generall had a higher winrate, but both top and midlane urgot dropped slightly that patch, only adc urgot gained a massuve 10% higher winrate.

RebonackMay 28, 20162 votes

In Diamond+ only with a play rate of less than 0.1%

Small sample size is small.

MizayaMay 28, 20162 votes

He's not bad honestly.

Lazypeon100May 28, 20162 votes

To be fair, Urgot isn't that bad. I actually do love playing him mis. Wouldn't dare take hkm as ADC though.

RookPusherMay 28, 20163 votes

He's a badass top lane.

Linna ExcelMay 28, 20161 votes

Doesn't matter as no one is playing him.

roflcopterssssssMay 28, 20161 votes

but urgot's not an ADC

Qiy4naMay 28, 20161 votes

http://champion.gg/statistics/#?roleSort=ADC&sortBy=general.winPercent&order=descend its true*[slayer-pantheon-popcorn]* http://champion.gg/champion/Urgot/ADC look at that win rate by patch and win rate % by games played

nice tableMay 28, 20161 votes

urgots win ratio goes up to 79% at 125+ games ...? this cant be correct ... http://champion.gg/champion/Urgot/ADC

Velzard of KozMay 28, 20161 votes

Not even a single "Urgot to be kidding me" joke was posted that day.

Cookimonster316May 29, 20161 votes

But also the lowest play rate so it's not that weird

KnalxzMay 29, 20161 votes

OMG URGOT IS NOT A BAD CHAMPION!

Alteast note after Riot updated his spells awhile ago. He's been powerful ever since and the ADC changes aka free bullshit only made him more powerful. IMO Urgot doesn't need a rework just a VU, the guy's fucking god tier.