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

< 10 Games 180 1.63%
11-40 Games 819 7.41%
41-80 Games 1,162 10.52%
81-120 Games 1,737 15.73%
121-160 Games 664 6.01%
161-200 Games 929 8.41%
200-399 Games 1,703 15.42%
400-599 Games 1,086 9.83%
600-799 Games 456 4.13%
800-999 Games 170 1.54%
1000+ Games 2,140 19.37%
Voters: 11046. 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.

ChaoticFool

Member

02-16-2012

Quote:
Originally Posted by Lyte View Post
I am interested in the idea of possibly presenting people with messages about their current match-up. I would not want to present people exact expected win rates based off Elos in the Loading Screen; however, there is potential for presenting messages such as, "You are Team Blue, the underdogs in this match!" or "You are Team Blue, the favored Champions in this match!"

The reason why I am hesitant to post exact expected win%s is because it is easy to misconstrue the meaning. If a matchmaker suggests that you have a 85% chance to win but your team loses, that does not mean the matchmaker necessarily failed. Afterall, if we created that specific match-up many many times, you would in fact be expected to lose 15% of those games. It is hard to interpret the 'success' or failure of a matchmaker based off 1 data point.
Starcraft 2 does this, telling you if they are slightly favored, favored, etc. during the loading screen.


Comment below rating threshold, click here to show it.

TheGreatWalk

Senior Member

02-16-2012

sent to your email as you requested:

also, why did I lose 42 elo that game, and I won 12 in the previous one? Isnt that a bit unfair, particularly with someone who ragequits at 1300+ elo?

EDITED BY LYTE :: Saved the screenshot, thank you!


Comment below rating threshold, click here to show it.

DJ Radix

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

Senior Member

02-16-2012

Quote:
Originally Posted by Lyte View Post
This has similar concerns to another poster's idea to give people more Elo for bigger margins of victory. There's a lot of room for people to game the system and stall games knowing that they can more Elo for it. If losers also lose less Elo, than both sides will want to stall
What about calculating margin of victory on a per-minute basis?

For example, calculate the following:

(Kills / Minute) * a
(Assists / Minute) * b
(Gold / Minute) * c
(Towers / Minute) * d
(Minions / Minute) * e
.
.
.
etc., where a, b, c, ... etc. are some constant weighting factors to keep all the values closely weighted. Gold is typically a much larger value than kills, for example, so Kills/Minute would need to be weighted up to measure against Gold/Minute.

Take the top 3 and bottom 3 values from that set. Use those values to calculate ELO gain. So in order to gain extra elo, you have to 1) be helpful in more than 1 way, 2) be helpful in a manner that is to scale with the game's length, 3) avoid being really really terrible in your worst areas.


Comment below rating threshold, click here to show it.

Dudoes

Senior Member

02-16-2012

I really appreciate you looking into how matchmaking works Lyte and be expecting to get your inbox spammed with screen shots starting tomorrow on my 3 day weekend


Comment below rating threshold, click here to show it.

DJ Radix

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

Senior Member

02-16-2012

Quote:
Originally Posted by Lyte View Post
Glicko systems have a 'confidence rating,' which is an extra parameter that represents how confident the system is that a player is currently at his true skill level. I think Glicko tries to solve the concerns you have raised.

Elo systems have K-factors, which are similar, but basically represent how important a match is. Early in a player's career, a K-factor starts high because the first few matches are important for determining the relative tier a player should be at.

League of Legends, however, has additional complexity which is why traditional Glicko or Elo systems are not sufficient. For example, a 1800 Vayne player is not necessarily a 1800 Annie player. This means that player performance and consistency varies greatly from game to game, adding a lot of additional variables that are hard to capture with traditional matchmaking systems.

In saying this, there are a lot of smart people at Riot, and we are going to do what we can to make the best matchmaker possible--it will never be perfect, but we will always keep working on it.
What if the K-factor varied based on the champion you were using? So the first time you play Annie ranked, there is a larger K-factor than if you are playing Vayne for the 40th time.


Comment below rating threshold, click here to show it.

Calipup

Senior Member

02-16-2012

I am at 1300-1400 elo after 80ish games. I feel like this really is my skill range at this moment. I mostly play support and jungle, never really have played the other 3 roles since I don't really feel like fighting for them.


Comment below rating threshold, click here to show it.

Hoggel

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

Senior Member

02-16-2012

Quote:
Originally Posted by NougetTime View Post
I havent been able to read through alot of this, but at 1800-1900 elo, the disparity of the people it pulls in is too large. For example, a good chunk of the time the team will be
1600 elo
1800
1900
2000
2000

Lets say the opposing team also has the same elo (which doesn't happen much at this elo). What ends up happening is that the 1600 elo will end up going against the 2000 elo, and guess what? You pretty much just lost the game. Its really frustrating. I'm short on time so sorry if this seems rushed, its because it is.
This is largely because of the thin amount of people above 1900 elo. There are over 100k people stuck in 1200-1500 and 1/3 that above 1700? People call this elo hell.


Comment below rating threshold, click here to show it.

BACKSTABUUU

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

Senior Member

02-16-2012

Hmm, the last imbalanced thing I've seen in matchmaking was a ranked game where the average elo of everyone was 1400, and there was a 1600 on the enemy team.

I didn't grab a screenshot though, so I'm sure it'd be a waste of time to email you.


Comment below rating threshold, click here to show it.

CupcakeTrap

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

Senior Member

02-16-2012

Quote:
Originally Posted by Lyte View Post
I'll double check this when I get back to Riot HQ, as I am currently at a convention. I am 90% sure of the answer, but want to double check first!
Best. Rioter. Ever. Thank you.

Quote:
Originally Posted by Lyte View Post
I am interested in the idea of possibly presenting people with messages about their current match-up. I would not want to present people exact expected win rates based off Elos in the Loading Screen; however, there is potential for presenting messages such as, "You are Team Blue, the underdogs in this match!" or "You are Team Blue, the favored Champions in this match!"

The reason why I am hesitant to post exact expected win%s is because it is easy to misconstrue the meaning. If a matchmaker suggests that you have a 85% chance to win but your team loses, that does not mean the matchmaker necessarily failed. Afterall, if we created that specific match-up many many times, you would in fact be expected to lose 15% of those games. It is hard to interpret the 'success' or failure of a matchmaker based off 1 data point.
Wouldn't allowing people to pick their desired win percentage lead to problems if choices were systematically skewed in one direction? I do like the idea, though. Different Normal Elo "profiles" would also be fun. I think it would be best in combination with a system for flagging your role prefs. e.g. a Veigar-mainer could flag "mage" as their role and use their "SRS BUSINESS" Elo profile.

It seems like this would be better than relying on MM to realize "Oh, CupcakeTrap is trying to play Riven now; better drop that Elo like a rock to find reasonable matches."


Comment below rating threshold, click here to show it.

Jixson

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

Senior Member

02-16-2012

I have a perfect example for your poor matchmaking. Last game I play.
Someone calls bot graves, someone calls mid, someone calls lee sin jungle. All good so far.I go top, playing Shen. Next pick...jungle gangplank. Does not speak forever in lobby or game. Person who called Lee Sin jungle then proceeds to pick jax(on "accident", said he "tried" to pick janna). He then proceeds to give up first blood,second blood, third blood. Trolls our team by stealing buffs from people who could actually benefit from it.

Yes the gp trolled us, but the guy who picked jax could have rolled with it,nah he wants to make us suffer for 39 minutes because he refused to surrender, as did the gp I think.

This is the stuff we need to clear out of the ranked game, and possibly the game completely.


edit: I just looked at the Jax from that game. 54 normal wins and 9 losses and a leave on his last 10 games page and no complete tier 3 rune page. We probably would have lost if he played jungle lee sin as well, as he has a 21% win rate with him, but he just picked jax to annoy us.