League of Legends: I finally updated the player list in the OP!

[SkyRiderX]

The MMR system is weird...I've heard of odd stories about it and the way it functions..

Also a huge apology for no being at tuesday's practice! Some urgent stuff came up and I spent the whole night on the phone and online dealing with stuff about my accident...

Another note: Our tournament is supposed to be this saturday however I havent heard much about how many teams will show up or much at all for that matter so I'll try to inquire more and keep you guys posted.

[kitty]

I haven't gotten a return email with the paypal info either.

[The Derrit]

Oh hey works pretty well in this game

[kitty]

Then go play DotA 2? The games couldn't be more different in what the developers emphasize. Saying that their matchmaking works for League is like saying I can slice an orange just like an apple.

Edited February 20, 2014 by kitty

[zircon]

Of that stuff, I thought the team gold earned was a particularly interesting one. Generally speaking, gold earned as a team correlates strongly to how well everyone is doing.

[kitty]

Team gold can be skewed very easily with champions like Mordekaiser or Gangplank who rack up a lot of gold with very little effort. Also, unlike DotA, you do not steal gold from the enemies that you kill in League of Legends. You can also deny in DotA and not in League. Considering these factors, it would be harder to draw up a reliable formula based on gold to calculate MMR in LoL than DotA.

[The Derrit]

Team gold can be skewed very easily with champions like Mordekaiser or Gangplank who rack up a lot of gold with very little effort. Also, unlike DotA, you do not steal gold from the enemies that you kill in League of Legends. You can also deny in DotA and not in League. Considering these factors, it would be harder to draw up a reliable formula based on gold to calculate MMR in LoL than DotA.

but not impossible by any means

look obviously i like playing league more or i wouldn't be here but at the end of the day, saying 'the only way (and the best way!) we can judge your skill is by whether you win or lose games that we are attempting to make you win 50% of the time' is laughable. there are obviously other ways. it is VERY likely that one of them is a better way.

i don't have some all-in-one solution but the current system could be improved and if you think the current system is the best it can possibly be, and that no improvements can be made so that MMR is a more valuable, objective statistic, you're fooling yourself.

[kitty]

I agree with you but the way you go about making suggestions is very undeveloped. Instead of saying "it works for this game!" try stating something about it. Just linking to DotA 2's patch notes isn't helpful and doesn't really add anything to the conversation.

[Nabeel Ansari]

Team gold can be skewed very easily with champions like Mordekaiser or Gangplank who rack up a lot of gold with very little effort.

Edited February 20, 2014 by Neblix

[zircon]

Team gold can be skewed very easily with champions like Mordekaiser or Gangplank who rack up a lot of gold with very little effort. Also, unlike DotA, you do not steal gold from the enemies that you kill in League of Legends. You can also deny in DotA and not in League. Considering these factors, it would be harder to draw up a reliable formula based on gold to calculate MMR in LoL than DotA.

True, it CAN be skewed, but it's pretty rare to see one player skew the *team* gold so much that it would appear like an even game when it actually isn't. Yeah, once in awhile someone will go 23-0 or something like that, but if they do then they're probably helping everyone else as well (ganks -> objectives / free farm). Surely the vast majority of the time, a game where one team has 22k gold and the other has 14k gold is more of a clear victory/loss than one where both teams are hovering around 22k.

You could even go a step further and look at if one player is skewing it and take that into account as well.

[Brushfire]

No inputs on the overlay?

[kitty]

True, it CAN be skewed, but it's pretty rare to see one player skew the *team* gold so much that it would appear like an even game when it actually isn't. Yeah, once in awhile someone will go 23-0 or something like that, but if they do then they're probably helping everyone else as well (ganks -> objectives / free farm). Surely the vast majority of the time, a game where one team has 22k gold and the other has 14k gold is more of a clear victory/loss than one where both teams are hovering around 22k.

You could even go a step further and look at if one player is skewing it and take that into account as well.

This is true, but they would also have to account for game length. Wouldn't you say that gold earned from champions killed early on is more effective than gold you got for clearing a couple waves that built up in end game? There are so many variables that go into gold earned that I don't think it's very viable as an indicator of skill.

No inputs on the overlay?

I thought it looked cool.

[relyanCe]

"it would be difficult so that means we shouldn't do it"

stop using this as a way to defend a shitty system from undergoing beneficial change

it's not on the community to figure out every nuance of how a thing could be made better, only that the community raises its voice and demands a reasonable, better way of handling what is currently a disgustingly toxic, negative-sum system

EDIT:

I don't want to spend another season picking champions that win to get my MMR up. It's dishonest and plain doesn't reflect on my progress as a player. It simply means I've broken down and integrated into whatever meta is enforced by which champion wins the most.

It's silly to think that I've become a better player because I main the S4 equivalent to S3 Ahri/LeBlanc, and if the game COULD measure the factors that actually go in to being a better player of the game League of Legends on an individual level for SoloQ (which it already does), instead of the enforced roullette of Win/Loss ratios, it SHOULD.

Aside, it's not that I don't think the current system doesn't work. It's just nowhere near as fun, rewarding, or honest to the game philosophy of LoL than it could be, and I wouldn't be surprised to see a huge departure of ranked LoL players to DotA 2 if it becomes clear they have a vastly superior ranking system.

EDIT EDIT:

Still no return email for the tourney on Saturday as well

Edited February 20, 2014 by relyanCe

[Bleck]

how is dota 2's ranking system not as broken as lol's

[ArmadonRK]

Oh, are we discussing the merits of the ranked matchmaking system again? I'll try not to repeat myself much, since my stance remains largely unchanged.

The ranked matchmaking system is neither efficient nor healthy to the competitive scene. Individual skill is only accurately charted if the skill gap between you and your peers is so high that you can carry your way out of elo hell.

Me? After trying ranked for about 150 games in Season 3 I gave up on that pretty much for good. It simply isn't fun being gated into a low elo by the sheer number of trolls and ragers because I'm not good enough to carry my way out. I still play ranked 5s, obviously, given my stance on premade teams. But solo, I exclusively play normals. It's a much more enjoyable experience, and my normal MMR is probably where it actually belongs, allowing me to actually focus on playing, having fun and improving, rather than the chore that is solo-queue. Not taking every match so seriously, not having every loss charted, not having to take a break every other match to avoid tilt.

That said, I now see the merits of the ranked matchmaking system that I didn't the last time I spoke on this issue. It seems that most the trolls, ragers, feeders, AFKers and smurfs are so busy in solo-queue that they don't have time to poison normal matches. While I used to believe normal queue was a very toxic environment, relative to ranked, it's a rather beautiful place.

All that said, if you guys want to know my suggestions for improving ranked matchmaking, I have none. League is, at the end of the day, a team game. There are no good ways to chart a player's progress outside of W/L. I see no way to improve the way player skill is charted. So, yeah, I still think ranked matchmaking is as much a joke as it has always been, and I don't expect that to ever change.

Okay, now that I have once again drawn your ire, feel free to add me in game.

Summoner Name: Shreyk

[Seven]

Again, this is going to sound kinda douchey, but if you think the current system is broken it just sounds like you think you 'deserve' to be placed higher than you actually are. Last season I raged and tilted and trapped myself in Silver. Over the preseason I learned roles other than support and played more relaxed when the season began, and now I'm GoldV. Maybe I can make Plat, who knows.

[Bleck]

Again, this is going to sound kinda douchey,

life hacks; if you know something is gonna make you sound like 'that guy', don't say it

[relyanCe]

also noting the faults of a system OBVIOUSLY means you have an entitlement complex OF COURSE

how is dota 2's ranking system not as broken as lol's

Dunno if it is or isn't or if it will or won't, but the team is willing to experiment with a new system that ostensibly will at least TRY to measure individual skill in a solo game setup and honestly anything sounds better than LoL soloq at this point

Edited February 20, 2014 by relyanCe

[ArmadonRK]

Again, this is going to sound kinda douchey, but if you think the current system is broken it just sounds like you think you 'deserve' to be placed higher than you actually are. Last season I raged and tilted and trapped myself in Silver. Over the preseason I learned roles other than support and played more relaxed when the season began, and now I'm GoldV. Maybe I can make Plat, who knows.

It's just simple, logical analysis. It's a team game. It doesn't track your skill, it tracks your win record. There's a mismatch there that means the lower you belong, the more likely you are to get trapped below your skill level.

I'm not going to pretend I'm some Diamond god. I don't even belong in Gold. But I'd definitely measure my current skills at somewhere in Silver, though they haven't always been. If I place in Bronze I, like I did last season, I expect to fall quickly to Bronze V and be trapped there. Because I'm not good enough to solo-carry all my games, and the way ranked is gated, that's where many of the most toxic players are trapped.

If I were a Plat or higher level player, yes, I'd be good enough to carry my way out of Bronze. But I'm not.

I have two RL friends. Both good team-players, of roughly equal skill, both tremendously better at this game than myself. One placed in Silver III last season, the other in Silver V. They both started off on unfortunate losing streaks. Player A dropped to Silver IV, then turned it around and made it up to Gold V. Player B dropped to Bronze I, and fell all the way to Bronze V.

Even though they were about the same level of skill, and I believe they both belonged in Gold, the one that got unlucky enough to end up in Bronze was trapped there.

So much of ranked matchmaking is luck, especially as you go lower in the standings. The real measure of skill will always lie with premades. Ranked 5s is actually about what League is about: where your team belongs in the rankings.

[Bleck]

Dunno if it is or isn't or if it will or won't, but the team is willing to experiment with a new system that ostensibly will at least TRY to measure individual skill in a solo game setup and honestly anything sounds better than LoL soloq at this point

trying doesn't actually solve the problem, especially if the goal is, like I said, impossible

I mean you can easily say that being stuck with someone significantly less 'skilled' than you are is going to be more bearable if you know in the back of your head that some irrelevant measure of effort has gone into making the matchmaking less bunk, but something tells me that everybody who says that is basically lying

[Seven]

life hacks; if you know something is gonna make you sound like 'that guy', don't say it

Definitely should've went with that one.

I think the larger issue at hand is that there are a lot of different definitions of what 'skill' is. If you feel you're more skilled than players you're matched with, then it just comes down to your definition of skill and whether you feel the matchmaking system correctly accounts for and represents your skill. Some players can CS really well and don't know what else to do, some players know how to duel in lane and get ahead, some players know how to teamfight etc. As you learn to do multiple things at a higher level on a more consistent basis, presumably you should move up the ladder (in the long run). I personally feel that this is the case at the moment.

[relyanCe]

trying doesn't actually solve the problem

actually, trying and experimentation does solve problems. see: all of science

especially if the goal is, like I said, impossible

explain why the goal is impossible without saying "It would be hard so we shouldn't try"

I mean you can easily say that being stuck with someone significantly less 'skilled' than you are is going to be more bearable if you know in the back of your head that some irrelevant measure of effort has gone into making the matchmaking less bunk, but something tells me that everybody who says that is basically lying

MMR isn't bunk. No one to my knowledge is saying it is. MMR works just fine for calculating players of similar skill as far as I've experienced.

I'm saying the way in which MMR is calculated and by extention the way LP is awarded in soloq is fundamentally misaimed and causes a great deal more frustration (which leads to greater player toxicity) than enjoyment (which would lead to less player toxicity). It isn't fun at all to play ranked soloQ, and how that has been allowed to basically remain unchanged is mindboggling.

EDIT: Reread your post and I think I got what you mean now. Thing is, people would be way less toxic about this kind of thing if your advancement as a player wasn't measured solely by winning. If it's okay that you are paired with someone less experienced or generally sucky, as long as you continue to play the game well on your own behalf and be rewarded for playing the game well, you'll probably be less likely to rage about how that person is ruining your ability to advance, since it's not all about winning.

I think the larger issue at hand is that there are a lot of different definitions of what 'skill' is. If you feel you're more skilled than players you're matched with, then it just comes down to your definition of skill and whether you feel the matchmaking system correctly accounts for and represents your skill.

I don't really think this gets at anything. Just because LoL players may define skill differently doesn't negate the fact that there are core skills (champion mechanics, CS, awareness, objectives, KDA) that rigidly define whether or not you are going to do well in the game of League of Legends

Some players can CS really well and don't know what else to do, some players know how to duel in lane and get ahead, some players know how to teamfight etc. As you learn to do multiple things at a higher level on a more consistent basis, presumably you should move up the ladder (in the long run). I personally feel that this is the case at the moment.

This is not the case. Well, no, it's part of the case, but not most of it. You move up the ranked soloq ladder in one of basically three ways:

1. You are immensely better than your placed tier, and will stop everyone until you get where you belong.

2. You get uncharacteristically lucky and win a large amount more often than you lose until you are placed somewhere where you realize you're still shit at the game but everyone else in the tier is less shit than you.

3. You game the current meta, take the champs that best capitalize statistically on the items/masteries/etc said meta, get good enough at those champs to regularly win lane, and carry games until you get into a tier where that won't work anymore, at which point you realize you're still shit at the game.

slowly working up your knowledge of the game in each aspect that is necessary for playing the game well will not reward you with steady upward movement in ranked. You will wallow, going up and down until you fall into one of the above three categories, because the game is designed to enforce a 50/50 winloss, WHICH IS NOT BAD. WHAT IS BAD IS THAT THE REWARD SYSTEM DOES NOT ENFORCE PROGRESSING IN SKILLS THAT BENEFIT BETTER LEAGUE PLAY. Instead, it prioritizes "winning" over every other aspect of League. So PEOPLE GO INTO RANKED TRYING TO WIN INSTEAD OF TRYING TO GET BETTER AT THE GAME. This makes sense for a 5-man team that functions as a unit and practices together, but makes NO SENSE in solo play.

League's reward system needs changing to make the player base enjoy the game more and genuinely feel like they're making process, which would work toward mitigating the immense toxicity of the community. Which is what I care about, NOT my ranking.

As an aside, it's been a great discussion so far. Thanks to everyone who's contributed.

Edited February 20, 2014 by relyanCe

[Bleck]

explain why the goal is impossible without saying "It would be hard so we shouldn't try"

It's impossible because it's not possible to separate individual contributions to a team's victory in a manner that doesn't arbitrarily define what individual actions are most important to a team's victory - i.e, it's not possible to determine who on a team did 'the best' without having a strict definition of what, exactly, 'the best' is actually supposed to be, and to have that Riot would essentially have to pare the game down to a binary contest of specific measurable values, thereby enforcing a metagame that rigidly opposes the concept of strategy and more or less fundamentally ruining the game.

The healthiest possible thing to do for the solo queue ranked system would be to stop pretending like it's actually possible to accurately rank skill levels for solo players - this is especially obvious when you consider that, for the most part, having a solo skill ranking system in the game is more or less meaningless anyway?

[Koriantor]

I haven't gotten a return email with the paypal info either.

I've been emailing them once every two or three days for the past week and a half and haven't heard anything from them.

...but yeah! Tournament soon. We should get an extra bit of practice in tomorrow night. Hopefully the servers aren't shaky like they've been all week.

[relyanCe]

It's impossible because it's not possible to separate individual contributions to a team's victory in a manner that doesn't arbitrarily define what individual actions are most important to a team's victory

What if I told you that the fact that achieving victory is literally the only decider of your ranked advancement in soloQ is part and parcel the source of all of soloQ's problems?

Why is the team's victory the only way in which we measure the ability of an individual in the game of League of Legends? This doesn't occur in any other team sport or team eSport. We wouldn't say that Faker or Piglet is a weak LoL player if SKT1 didn't win Worlds, would we? Why use that mentality in soloQ? If soloQ is designed to reflect individual ability at LoL, why is advancement 100% contingent on the actions of a team? You do realize this concept is gamed mercilessly in soloQ, and soloQ games that are won by team tactics are an extreme rarity, right?

- i.e, it's not possible to determine who on a team did 'the best' without having a strict definition of what, exactly, 'the best' is actually supposed to be

Would you agree that there exists a set of things that a LoL player should do as opposed to other things to play the game better and advance in skill as a player? If you would agree, then you've admitted that 'the best' is definable and measurable. I would say there exists such a set of skills.

, and to have that Riot would essentially have to pare the game down to a binary contest of specific measurable values, thereby enforcing a metagame that rigidly opposes the concept of strategy and more or less fundamentally ruining the game.

Would you agree that the concept of strategy in LoL is reducible to actions taken within the game world? If so, would you agree that all actions taken within the game world are measurable and recordable (and in fact are)? If so, why would you think that measurable actions taken within the game world (which the concept of strategy supervenes upon) rigidly oppose the concept of strategy? Edited February 20, 2014 by relyanCe