How many games do you expect to play before you reach your true Elo?

< 10 Games 33 1.79%
11-40 Games 111 6.03%
41-80 Games 155 8.42%
81-120 Games 233 12.66%
121-160 Games 117 6.36%
161-200 Games 147 7.99%
200+ Games 300 16.30%
400+ Games 225 12.23%
600+ Games 90 4.89%
800+ Games 55 2.99%
1000+ Games 374 20.33%
Voters: 1840. You may not vote on this poll

Help Riot improve matchmaking! Looking for examples of bad matchmaking

First Riot Post
Comment below rating threshold, click here to show it.

Moyuba

This user has referred a friend to League of Legends, click for more information

Senior Member

18-02-2012

i was matched with a level 15 warwick yesterday or the day before. he went 0/7 in lane to a cho gath despite me baby sitting him.

it was a lot of fun.


Comment below rating threshold, click here to show it.

bobTHEsnob

Senior Member

18-02-2012

Quote:
Originally Posted by Comornic View Post
elo isn't you success rate with a given champion
elo indicates where you are OVERALL as a solo queue player
Well...that's my point(duh?). You may be super skilled with some champs(or roles) you play and suck with others...hence leading to many complaints since said player will pwn half his games and feed the rest. An effect of that is extreme variance for players and causing the number of games needed before players hit the elo they should be at to rise significantly. Now...you could say work as intended since clearly the game requires different roles/champs...BUT I would claim it is not, since this situation leads to a lot of frustration among the players.


Comment below rating threshold, click here to show it.

Sabotejcz

Junior Member

18-02-2012

I think if u played more than 100 or 200 games you shouldnt be put in the same team with a guy that played 20 games. Despite hes at the same ELO as you are. And its stupid that you only get bonus for first games, which lead to either ur very lucky, got good teams and you put urself at 1600-1800 elo, or u got trolls and after 50 games youll be at 1200 or 1300. Make every game gives the same amount of elo like 12-13 if players are much higher then like 20 or so, and it should be fair.


Comment below rating threshold, click here to show it.

Pyrrian

Senior Member

18-02-2012

Quote:
Originally Posted by Comornic View Post
In the other topic I suggested an experiment concept which has to be carried at the start of a ladder season:

perhaps you guys can use it, improve some points and see how it goes
What do you try to prove with your experiment? I bet all these pro players will be in the top 100 in no time.


Comment below rating threshold, click here to show it.

i came

This user has referred a friend to League of Legends, click for more information

Senior Member

18-02-2012

Actually as soon as I hit lvl 25 on my account I only played with 30's. I have no idea why, I didn't queue with any of my friends but this still happened in like 4 out of 5 games. Was really pissed at that time.


Comment below rating threshold, click here to show it.

Comornic

Senior Member

18-02-2012

Quote:
Originally Posted by Pyrrian View Post
What do you try to prove with your experiment? I bet all these pro players will be in the top 100 in no time.
thats the difference - you bet and I want proof...

to prove the flaw of ELO system, you have to assume it's correct

choose 10-20 top elo players and give them 10-20 clean accounts with random nicknames where noone will recognise them.

if by the mid-end of the season their ELO will be significantly different from the one they have now - means only one thing:
ELO system is flawed, because when you see those players play and justify their choices in games, you have no doubt that they are experts and you have nothing but respect for them.

so you don't have to be so aggressive towards ME - I'm not trying to prove that "progamers suck and I am awesome".

I'm trying to prove there is a flaw in the system that prevents from correctly rewarding people for their performance in game

1st step is to prove it exists
next is to find a way to fix it.
and then ... fix it ;-)

the entire LoL community is build around solo queue, so it is pretty important to make it fair for everyone.


Comment below rating threshold, click here to show it.

Hachiman

Junior Member

18-02-2012

Well i canīt really send and actual screenshot but this is what happens too me all the time :

I play with 4 friends, normal not ranked

Me around 890 wins something around 50~51% win/loose ratio ( ranked 1525)
Friend 1 around 880 wins 53%
Friend 2 around 850 wins 50~51%
Friend 3 around 200 wins 50~54% (yes we carried him hard too Lev 30 ^^)
Friend 4 around 300 wins 50%

as you can see there is a huge difference in played games but the win/loose ratio is always around 50% (donīt know the exact numbers) and our Problem is this :

a) The game is sooooo easy we totally dominate the game and its an way too easy victory
Enemy team is then at a maximum of 500 wins but there are almost all below

b) we get crushed, i mean completly crushed, all lanes are failing and we have no chance of winning
4 of 5 players have around 900~1000+ games played and 1 is at 200 wins

Donīt get me wrong i know matchmaking doesnīt count games played, but the thing is there is no way (expect for smurfs) that a player with 300 wins or 400 wins has the same amount of experience and rune pages with runes (!) as somebody with 800+ wins ....

I think the biggest problems is that "amount of games played" isnīt part of the matchmaking system since this is a good indicator for experience

Sadly i canīt remember the last time i had a close 50+ min game where every team could win if the the enemy makes 1 mistake ...


Comment below rating threshold, click here to show it.

Oddland

Senior Member

18-02-2012

**** that game I had a week ago would have been perfect for that: went 21/6/x with Tristana, had 2 pentakills in one game and it still wasn't enough to win us the game because the other players in my team were so bad.


Comment below rating threshold, click here to show it.

I love Kurisu

Senior Member

18-02-2012

Quote:
Originally Posted by Riot Lyte View Post
I think duo queue is an interesting problem. Sometimes if the duo queue players are close in Elo, they can control for more positions in a game and therefore have an easier time up the ladder; however, if the duo queue players are far apart in Elo, I believe they actually have a harder time climbing up the ladder because the carry in the game cannot overcome the feeder effect in the game. I am investigating these issues and I'll see what the data says.
The problem with duoqueue is that you'll end up with a generally lower elo team than the enemies.

The difference of 80-150 elo can mean a lot, and usually determines who loses laning phrase and then proceeds to die.

Yes, sometimes duoqueue makes it easier rising in the ladder, it still screws up with matchmaking tho.


Comment below rating threshold, click here to show it.

HMHS8

This user has referred a friend to League of Legends, click for more information

Senior Member

18-02-2012

You should consider adding more stats and match players with similar results. for example:

4 players from team "X" get average:
- "X" minions per min / game
- "X" wards (if player was punished by trolling in game "Y" decrease his average ward count by "X") - this should help deal with those who troll and spam wards around

so dont give those 4 players newbe who can't even last hit nor ward properly and dies 10 times every game.


I dont care IF it kills 0 CS support meta, just give me an option to avoid players who are much worse on last hiting and warding than me. It's really frustrating when you are the only person buying wards in your team, and you are not even playing support.

Also normal matchmaking need an option to avoid getting matched with low lvl's. I don't care if they duo q.

Attached Thumbnails
Click image for larger version

Name:	real matchmaking.JPG
Views:	179
Size:	68.6 KB
ID:	75855