Graves jumped from a 50-52% winrate to 55-56% in every role

Decrit·Jan 18, 2016·27 votes·9,660 views

... without receiving significant buffs.

The only changes he got were indirect ones, that affected some niche scenarios ( matchups mostly, like the Cait bugfixes or the trundle nerfs ), the Last Whisper change ( that affected equally all the users ) and then lots of buffs that did not affect him but his competitors, making the enemies more powerful.

Still he got this significant change both in jungle and in lane.

What does it mean? Probably the sample size is small, probably during the prior patch the winrate increased during the long month meaning he had a way lower winrate early in the month and a much higher one later ( but still is an absurdly big change), or probably just something else.

What i do want to say is the old tale of "do not post statistical research as a way to support your opinion, use it to make things clearer once those are investigated properly".

73 Comments

RiotMeddlerJan 19, 201625 votes

{quoted}

What i do want to say is the old tale of "do not post statistical research as a way to support your opinion, use it to make things clearer once those are investigated properly".

+1 to that. We think win rate's a useful metric, but it's an indicator that something needs to be looked at, not something that should decide things by itself.

In this particular case that sort of spike in win rate certainly means Graves is worth a closer look. From a quick glance at our internal data it looks like some of that stated increase is low sample size or a sampling issue (we're seeing him up a bit, but nowhere near 4%). Looking at his files there don't seem to be any significant changes that weren't in the patch notes either (couple of visual bug fixes for skins, but that's about it). Good champ to keep an eye on for a couple of weeks as a result, see if there's a trend going on or just a temporary spike.

The Blue RajahJan 19, 20169 votes

A good, solid, sensible reply. Thank you.

manbearswineJan 19, 20164 votes

[{quoted}](name=The Blue Rajah,realm=NA,application-id=3ErqAdtq,discussion-id=6Jzlocg6,comment-id=000e0000,timestamp=2016-01-19T07:33:28.004+0000)

A good, solid, sensible reply. Thank you.

Albeit inconsistent with plenty of statements claiming winrate=strength.

Meh,go figure.

BetaDude40Jan 19, 20164 votes

Questioning if you even read the post...

manbearswineJan 19, 20164 votes

{quoted}

?

Questioning if you even read the post...

What exactly don't you get?

His reply states winrate by itself might only imply that something is there,but doesn't necessarily say anything.

Other replies from reds openly stated that winrate=strength,what exactly don't you understand? Do you not see the contradiction here?

Stiv1234Jan 19, 20163 votes

The way your post is worded it suggests that you see contradictions within his post: "it's inconsistent (and) with plenty ...". I assume you meant contradictions with former posts of Reds. Maybe you should clarify that a bit.

manbearswineJan 19, 20162 votes

{quoted}

The way your post is worded it suggests that you see contradictions within his post: "it's inconsistent (and) with plenty ...". I assume you meant contradictions with former posts of Reds. Maybe you should clarify that a bit.

Well I figured it was obvious that that was the case,being inconsistent with something a bunch of random players stated would be kind of a pointless thing to point out.

Photos of GhostsJan 20, 20161 votes

Read again

manbearswineJan 20, 20161 votes

[{quoted}](name=Photos of Ghosts,realm=EUW,application-id=3ErqAdtq,discussion-id=6Jzlocg6,comment-id=000e0000000000000000000000000000,timestamp=2016-01-20T01:24:38.856+0000)

Read again

I still don't see it.

Photos of GhostsJan 20, 20161 votes

{quoted}

The way your post is worded it suggests that you see contradictions within his post: "it's inconsistent (and) with plenty ...". I assume you meant contradictions with former posts of Reds. Maybe you should clarify that a bit.

within his post

his

Not between his post and other people's.

manbearswineJan 20, 20161 votes

[{quoted}](name=Photos of Ghosts,realm=EUW,application-id=3ErqAdtq,discussion-id=6Jzlocg6,comment-id=000e000000000000000000000000000000000000,timestamp=2016-01-20T01:36:29.356+0000)

Not between his post and other people's.

I'd say there's a pretty obvious contradiction between his post and zenon's post.

Photos of GhostsJan 20, 20161 votes

You MEANT that. But what Stiv said is that your comment sounded like you meant there were contradictions within Meddler's post. And upon Stiv telling you that, you answred it was obvious that Meddler was going to contradict the opinions of random people on the forums. Hence, you failed at reading Stiv's comment. Hence, read again.

manbearswineJan 20, 20161 votes

[{quoted}](name=Photos of Ghosts,realm=EUW,application-id=3ErqAdtq,discussion-id=6Jzlocg6,comment-id=000e00000000000000000000000000000000000000000000,timestamp=2016-01-20T12:26:44.013+0000)

You MEANT that.

Ofc I meant that,what else would I mean.

But what Stiv said is that your comment sounded like you meant there were contradictions within Meddler's post.

How could that possibly be the case? All I said was "Albeit inconsistent with plenty of statements claiming winrate=strength."

Does this look like the phrasing one would choose for contradictions within the post itself? Wouldn't I just say that post contradicts itself then? Instead of just "statements"?

And upon Stiv telling you that, you answred it was obvious that Meddler was going to contradict the opinions of random people on the forums.

Basically the idea was: "Well obviously I wasn't talking about it being self contradictory to the post on it's own,hence it's either 1)contradictory to posts made by a bunch of random people 2)contradictory to other posts made by reds"

I then went on to say that the first point isn't worth making,hence it's obviously option number 2.

Mind you I didn't go to such details explaining what I was getting at,as I thought it to be obvious,but clearly I keep overestimating the people here.

Hence, you failed at reading Stiv's comment. Hence, read again.

I read his comment,I just didn't expect such simple-mindedness,but fine,thank you for clarifying this to me,I'm just not used to dealing with people with such...a way of thinking.

RiotMeddlerJan 19, 20165 votes

Do you have a post in mind for the 'winrate=strength' comment by any chance? That's a subject I think it'd be great if we could talk about in more detail.

Lord AzırJan 19, 20161 votes

{quoted}

Do you have a post in mind for the 'winrate=strength' comment by any chance? That's a subject I think it'd be great if we could talk about in more detail.

It is true though isn't it? Skarner was at a what 65% win rate when he was first reworked.

RiotMeddlerJan 20, 20164 votes

Win rate's a useful indicator, and particularly extreme numbers are generally only found where there's a significant power problem (main exception: new champions/reworks people are still learning). I don't mean to say that win rate doesn't give some cues as to a champions power in the slightest, but it's just one piece of the discussion, especially when the differences from the norm are pretty small. Le Blanc at a 49% win rate has historically been pretty out of line for example, despite being so close to 50%, because of how effective she is when played well (experienced players and into the right match ups).

LankPantsJan 20, 20162 votes

Le Blanc at a 49% win rate has historically been pretty out of line for example, despite being so close to 50%, because of how effective she is when played well (experienced players and into the right match ups).

Even look at Lee Sin. He's probably too powerful now and he has a 47%ish win rate.

I think for a lot of champs their winrate correlates more to difficulty than actual balance or power.

RetillinJan 20, 20161 votes

Touché

AyRe CoNteMpTJan 20, 20162 votes

"difficulty" is such a strong word. dont use it for lee sin, hes not difficult, he just requires some "getting used to"... but in general, you are right. tho you shouldnt expect rito to understand any of it because he can kick.

HellioningJan 19, 20163 votes

This post by Zenon straight up says, and I quote,

Power is measured by win rate. Obviously you then have a bunch of conditionals to that winrate (what ELO are we looking at? How does this behave over time? How does the winrate change with game length? What positions are we looking at?), but in the end of the day, power is knowable.

LostFr0stJan 19, 20162 votes

Hello Meddler,

I know this comment doesn't belong here, but I'm not sure who to ask and it's a pretty important question, so here it is:

There is currently no stickied post for Shen feedback on the pbe.
We have had no luck the past week contacting the pbe reds on this.

Would it be possible to have this thread http://boards.pbe.leagueoflegends.com/en/c/champions-gameplay-feedback/iIEyZa0n-master-shen-euwtoplane-the-most-complete-and-explicative-feedback-about-the-recent-shen-update stickied to the Champions & Gameplay Feedback section?

(This thread has become the current hub for Shen-related discussion; we are trying very hard to keep it on the first page for easy access.)

Thank you for understanding, and I am sorry for intruding on your winrate discussion.

Sincerely,

LostFr0st

manbearswineJan 19, 20165 votes

from here: http://boards.na.leagueoflegends.com/en/c/developer-corner/HZEhZgif-followup-tahm-kench

7th red post by ZenonTheStoic

Let's talk about power vs frustration for a moment!

You start out your post by saying that Tahm is over his power budget and then listing a number of ways that his spells do too much (and you're onto something for most of these, actually; they're all tuned very strongly).

In this thread I'm looking at frustration much more than power. Power is easy to measure; frustration isn't. Power-wise, Tahm is actually in a good spot (pre-season may have given him a slight bump, but he was very weak before).

**Power is measured by win rate. Obviously you then have a bunch of conditionals to that winrate (what ELO are we looking at? How does this behave over time? How does the winrate change with game length? What positions are we looking at?), but in the end of the day, power is knowable. ** Internal tools show Tahm Kench at a 47-48% winrate in the last three weeks across all regions and game lengths, in Normal 5v5s, Ranked Solo 5v5s, and Teambuilder 5v5s at ELOs of 1500+. If we expand to all ELOs, I see him at 49%, but at a slight downward trend (suggesting that at lower ELOs people are doing a worse job playing against Tahm, but are learning). Pre-season bumped him up by about 2%, so he was at 45-46% before in high ELO.

So again, if you're arguing that his power is too high, I will politely disagree, which is why I phrased this thread not in terms of power, but in terms of frustration. Frustration can come from power, but I don't think that's the case here. Janna has had 53/54% winrates literally for years now, and I never hear people talk about Janna as a highly frustrating character to play against (although Gosu may disagree with me here. Is Janna powerful? Absolutely. That E alone is ridiculous, and she is the queen of disengage. But is she frustrating to play around? I don't think so.

There were some other posts,but I think that this one started the notion that winrate=power,at least as far as red posts go.

RiotMeddlerJan 20, 20163 votes

Thanks for the link, I'll follow up with Zenon. Win rate's a useful indicator, but certainly not one we should be describing as directly equal to power or acting as though it is. We'd certainly expect a strong correlation between the two, but not an equivalence.

RiotMeddlerJan 20, 20162 votes

Oh, and for anyone interested in longer thoughts on how we approach data and champion balance there are a couple of good articles that Jules and Jag wrote recently on the subject here:

http://na.leagueoflegends.com/en/news/game-updates/gameplay/data-and-champion-balance-part-1 http://na.leagueoflegends.com/en/news/game-updates/gameplay/data-and-champion-balance-part-2

Utility VeigarJan 19, 20162 votes

THANK YOU. It's like when people say that Heimerdinger is good because he has a high winrate. He really isn't. He doesn't see tournament play for reasons.

ProfDrDeathJan 19, 20161 votes

If you don't mind the question:

What do you guys at Riot think about the performance of the new AP jungle item? Do your internal data confirm champion.gg's observation of it massively raising the win rates of the AP junglers?

Earl EulrichJan 19, 20165 votes

so...is that what you did before nerfing Swain, while leaving Ryze all broken at the same time?

I mean...i would say what you describe is the right approach here, i just can´t see any of that approach in recent balancing.

Xela SyabJan 19, 20162 votes

You should lane against a Graves, it's pretty apparent why he's being played more and getting more wins. He's pretty oppressive and mobile which is why he's good in any lane and jungle. Also due to his rework he doesn't fall off like he used to and is a significant threat at all stages of the game (prenerf Lucian). Since you guys finally nerfed MF he can finally do his thing without being kite stomped. There is also some bad clarity with his ult, it seems to hit you in weird ways or just completely miss at point blank.

Akali is SO HOTJan 18, 20165 votes

???????????

They put his Q on his auto attack and he can get 120 armor and magic resist for free. They removed Akali's armor and magic resist from her shroud because it was OP but then gave a ranged assassin more than double that right after lol.

DecritJan 18, 20167 votes

...and?

i am not talking about what the rework did, that happened few patches ago already.

DroogzyJan 18, 20165 votes

I would say its most of an increase in popularity since he was seen in LCS, I've played him a few times in jungle since seeing him on LCS scene since I was curious to see how it was.

MrBuffingtonJan 18, 20162 votes

Usually an increase in popularity drops a champion's winrate, because an influx of players who don't know how to play the champion start coming in, which drags the average down from the mains who are actually good at the champion.

TotalJerkJan 19, 20164 votes

The LCS popularity probably showed a lot about how to properly build and play Graves.

DroogzyJan 19, 20164 votes

{quoted}

The LCS popularity probably showed a lot about how to properly build and play Graves.

Yeah I think that's what really influenced it, there were a lot of different builds and ways to play Graves, no one was really sure how the best way to play him was, outside of the fact that he IS a viable choice in the jungle. Now that build is clearer, it might be easier to win with him.

AyRe CoNteMpTJan 18, 20163 votes

its because graves is a champion who can deal alot of damage in a very simple, braindead manner and without putting himself in danger too much. this means, its hard to "fuck it up" with him so noobs dont lower his winrate by their noobness. the point about graves is, you have to respect his damage and since league players tend to respect nothing and nobody and forget everything they learned from the last game instantly, he owns people pretty hard.

oh and also, the instant double Q is too strong.

DecritJan 18, 20168 votes

Tossing aside how much of your reply is very subjective, consider that these % come from champion.gg, which analyzed only games from plat+

Now, plat players can screw up a lot too, but is far less convincing that there is a scenario as you say.

AyRe CoNteMpTJan 18, 20162 votes

im too lazy to check your elo but you are probably silver so i can understand that from your point of view, platinums are allready really good. let me tell you tho, thats not the case. the mentality and intelligence of players actually doesnt change throughout the elos.

Shiyo KozukiJan 18, 201617 votes

You shouldn't be commenting on anyone's ELO when you have an alternating caps user name.

AyRe CoNteMpTJan 18, 20163 votes

yeah because thats correlated

but whats correlated is peoples intelligence and the amount of judgments people make about other people based on superficialities.

also, seriously, you trashtalk me because of my name while your own is just some random generic japanese sounding anime name? [zombie-brand-clap]

BetaDude40Jan 19, 20163 votes

[{quoted}](name=AyRe CoNteMpT,realm=EUW,application-id=3ErqAdtq,discussion-id=6Jzlocg6,comment-id=00000000000000000000,timestamp=2016-01-18T18:36:50.661+0000) but whats correlated is peoples intelligence and the amount of judgments people make about other people based on superficialities.

Hmm... That sounds familiar...

im too lazy to check your elo but you are probably silver so i can understand that from your point of view

AyRe CoNteMpTJan 19, 20162 votes

well names and rank are definitely not the same. a nickname is something you chose at some point in your life for whatever reason. a rank is something you achieved, something you worked for. its like when i see you building a hut but it looks like this https://c1.staticflickr.com/5/4144/5066671755_ae751220dd_b.jpg

while mine looks like this http://i.imgur.com/3V94Jqc.jpg

i can safely assume that you are probably not the greatest expert at building mud huts, even if you got a japanese anime name thats really creative.

BetaDude40Jan 19, 20162 votes

Yeah, but did you go inside the huts? Do you know how long it took for each hut to be built, or which one is more structurally sound, or which one is capable of surviving natural disasters more effectively, or anything other than what they look like from the outside?

That's what I mean. You can't judge someone's construction skills purely off of what you see from the outside. Sure, the second one might look prettier, but how do you know that hut is cramped all to hell on the inside?

This is similar to this game. You can't judge someone's skills based off of how pretty their loading border is. Now, there's people who probably have Taj Mahals in comparison to these mud huts, which are superior in every way. But the average hut-builder doesn't have the time or resources to make a scale replica of the Taj mahal, he just wants a hut to sleep in for the night. When it comes to the vast majority of us, who don't have the privilege to build things on such a grandeur scale, people may think what we build is worthless. But you can't just take the mud hut from face value, you know. The pretty outside (or loading border or ranked icon or status or whatever) doesn't equate to one builder knowing more than the other. And, in essence, if you believe that your hut is better because it looks prettier (or any other analogy you can throw down), you're being judgmental, and you should stop worrying about how shit other huts look from the outside and instead worry about how shit yours might look from the inside.

Jesus that was a long post...

AyRe CoNteMpTJan 20, 20161 votes

well, even tho there is a slight chance that your hut is littered with diamonds from the inside, a silver player is a silver player and a challenger player is a challenger player. the silver is simply very extremely unlikely on a level where he can even understand what the challenger player saith because his own view of the game is so vastly different. he doesnt just do what the silver does but better, he percieves the game differently, he thinks about the game differently and he approaches it differently.

but all that aside, ofcourse im being judgmental because i believe that peoples ranks have a meaning while you are trashtalking me because of my name ; )

BetaDude40Jan 20, 20162 votes

I never made fun of your name, that was Shiyo.

And I think ranks only have a meaning in the form of a little consolation prize. I'd gladly fight a Challenger tier player, and while I don't know if I'd win or not, I wouldn't enter it thinking I was screwed or that I somehow know more about the game than him. If he wins, fine. If I win, cool. Everyone is just another player to me in this game; rank factors nothing to how I play. Being Diamond doesn't magically make you more skilled than everyone else (as we all know from boosted accounts). There is, admittedly, a correlation between skill and rank, but honestly not enough to warrant people being dicks to low elo players like we can't understand the game.

Like really, why do people think we're mentally handicapped or something?

AyRe CoNteMpTJan 20, 20161 votes

oh yeah right with the name im sorry, i get like 20 replies per day and mustve confused this. my bad.

with the ranks, its really cool that you dont have a mindset of defeat when facing a challenger or high elo player/team in general.

and OFCOURSE a rank doesnt mean shit by itself because you could be boosted, accbuyer or whatever but if that rank is genuinely earned, then that MOST LIKELY places you somewhere around that skill level. even tho this is only your SOLOQ prowess and doesnt neccessarily translate into broadband game knowledge. you can be a shaco main with 1000 games and diamond and still dont know jack shit about the game except how to cheese noobs. tho what is for sure is that you know better how to cheese noobs than a shaco main stuck in gold.

just in general i'd really wish that people would respect ranks more. they DO have a meaning and ofcourse they dont incline you to be a dick but it just is very hard not to be when you meet so much disrespect. literally people asking for advice but what they want to hear as an answer has to fit their EXTREMELY LIMITED LOW ELO view of the game.

i mean check out my first post. "its because graves is a champion who can deal alot of damage in a very simple, braindead manner and without putting himself in danger too much. this means, its hard to "fuck it up" with him so noobs dont lower his winrate by their noobness. the point about graves is, you have to respect his damage and since league players tend to respect nothing and nobody and forget everything they learned from the last game instantly, he owns people pretty hard.

oh and also, the instant double Q is too strong."

sure it might not sound super friendly but its a very valid explanation and it is also the truth but it has 22 downvotes becauses you know, noobs will be noobs.

magic poleJan 19, 20163 votes

why this is so up voted? what does that have to do with anything

NekusenJan 18, 20162 votes

Another proof that winrates mean nothing.

DecritJan 18, 20166 votes

It' s not much about that, is more about how they are used poorly. By themselves they mean nothing, because you can twist the interpretation as you want, but when in a context they can give clarity