### What is ELO hell?

TwistWrist

Senior Member

Quote:
Originally Posted by SargentCannibal
In games where an individuals Elo rating is determined by the success of the team that they are on and not on theiir individual accomplishments, Elo hell is not just a theoretical happenstance, it is a proven mathematical phenomenon.
Actually, math would prove the opposite.

You and 4 Elo hell players against 5 Elo hell players.

If you are truly better, then you should win in the long run and pull yourself out.

Only answer if you are still stuck there is that you aren't any better.

Panzerfaust

Emissary of the League

Quote:
Originally Posted by TwistWrist
Actually, math would prove the opposite.

You and 4 Elo hell players against 5 Elo hell players.

If you are truly better, then you should win in the long run and pull yourself out.

Only answer if you are still stuck there is that you aren't any better.
Aww you beat me to it. I was going to ask him where he got his math degree. Then I'd go ask one of my math professors about it (which I still may ).

Xathoz

Junior Member

Sorry if this has already been stated. I didn't read everything :P, but wouldn't there be other people who got into "elo hell" and are trying to get out. So, if you actually are good and are only there due to bad luck, then wouldn't you destroy other people and get out easily? Also, about the feeder excuse, wouldn't there be feeders on BOTH teams?

Xathoz

Junior Member

Wouldn't there be other people there due to bad luck or whatever your excuse is? So, it wouldn't actually be hard to get out. If you are saying feeders is the excuse wouldn't feeders be on both teams? Meh, that's my opinion. Sorry if someone already said this.

Gibgezr

Senior Member

In the long run, there is no elo hell. The thing is, the LONG run can be....long.

A statistician will tell you that in the short run, elo hell exists, and yes, it can be mathematically proven. He might draw you a dull graph, with that distinctive bell-curve shape, and tell you that the graph represents the balance of the teams "true" skill ratios in solo queue over 50 games. In the middle of the graph are the large percentage of the players whose ratios approached 1:1 i.e most of the time their randomly assigned teammates where about as good as the opposing team's players. These people never experience "elo hell".

On the far side of the graph is a much smaller area, just a few percent, where the players exist who had fortune smile upon them, and their first 50 games saw them getting a propondurance of mismatchs in their favour. They never experienced "elo hell".

On the near side...uhoh. There are the small percentage of players who had the run of bad luck. They got matched way too often with poorer teammates than the opposing team. They probably lost a lot of their first 50 games as a result of this.

In the short run (short being 50 games!), they had bad luck and their elo dropped rapidly. Now, they need to have at least "even" luck, and start digging their way out of the elo hell and get to matches that feature players more on their true skill level. Of course, it takes a lot more games to get the elo up now than it did to lower it initially. Thus...elo hell. They spend the next 150 games trying to slowly reach their "true" level, and thus play with people of equal skill.

This doesn't happen for the majority of players. It only happens at all in LoL because it is a team sport: in chess, a beginner's elo may jump around a lot in their first ranked matches, but it will quickly stabilize. There is no "bad luck" because there are no randomized teammates.

So, yes, "in the long run" everyone will reach their proper elo. For a small percentage of the player base, that "long run" will take a lot of games. I think Riot has done an excellent job of designing a ranking system that tries to keep things as fair as possible. There isn't anything for people to get scared about, but saying "there is no elo hell" is equivalent to saying "there is no randomness involved in 1200 elo player matchups" and "elo ratings points awarded per win/loss don't scale down as you play more games", both of which are patently untrue. The best you can say is "most people will not experience elo hell".

timosha161RND

Junior Member

.

TwistWrist

Senior Member

Quote:
Originally Posted by Gibgezr
There isn't anything for people to get scared about, but saying "there is no elo hell" is equivalent to saying "there is no randomness involved in 1200 elo player matchups" and "elo ratings points awarded per win/loss don't scale down as you play more games", both of which are patently untrue. The best you can say is "most people will not experience elo hell".
Depends what you consider Elo hell.

Elo with bad players? Of course that'll exist.

Elo where good players can get stuck for a long time? No.

Plenty of "pro" players have done streams where they go from a really low elo to a relatively high elo. I don't believe those generally takes months.

Junior Member

Elo hell exists, playing akali 43 ranked matches, 15 wins with an average score of 22/7/18 stuck at 800 tell me what the **** is up with this ****ty matchmaking system

Panzerfaust

Emissary of the League

Quote:
Elo hell exists, playing akali 43 ranked matches, 15 wins with an average score of 22/7/18 stuck at 800 tell me what the **** is up with this ****ty matchmaking system
A high K/D ratio does _not_ mean you were helping your team win the game. Kills don't win game, destroying the enemy nexus does. Kills only make it a bit easier to do so, if you're already doing the right things (i.e. pushing when you need to be, killing towers, etc).

" 'I am losing' => 'Matchmaking sucks' " is a false statement, please don't use it as if it were true.

KirbyCake

Senior Member

Mathematically and statistically, elo hell does happen often in the short run. However, if you play enough games and you are good enough, then you should be able to pull yourself out of elo hell.