Boards›Gameplay

Is individual performance tied into rank calculation in any way?

Karunamon··1 votes·314 views

We've all had a game where your own team got just completely roflstomped. Whether it be bad composition, an early mistake leading into a snowball, whatever.

Occasionally though, during one those games, one or two people really shine. Getting all the ganks, making all the right calls, never getting caught out, ending with a positive KDA and an objective or two when most everyone else has been just absolutely dumpstered.

And yet, as far as I can tell, they'll eat that loss just like everyone else will. In this particular case, this one player was saddled with a losing team.

This seems.. at the very least, unintuitive. An individual's performance is only one-fifth of the result of the game, yet their individual rank is tied to the game result, not their own performance. I can see how this would lead to wildly imbalanced matchmaking, where underperformers wind up above their true skill (carried by awesome teams), and overperformers wind up below it (held back by bad teams).

I can't believe that this isn't factored into the calculations somehow, but I can't find anything that suggests how it works.

Is it?

10 Comments

xPhantomDancerX2 votes

It does, but not in the way you think. If you do really well, riot saddles your team with under performers so you can prove you can actually "carry". For example, I'm a s2 players, but in my last game I got an unranked on my team, and on my second to last game I got TWO unranked and a bronze on my team, while in both games, the enemy teams were ALL silver players.

Smitty Baccal1 votes

It isnt calculated. Think about it that way, if a person outperforms everygame in the long run that person will eventually win more % of their games than their <<underperforming>> allies. Thoses person who SEEM to be better than the elo they are in most likely will throw future games or be unconsistent in their plays.

Mephisto31 votes

This is the number one "myth" that people who do not understand probability theory and matchmaking basics always believe in... what you are explaining is the following: the system assumes you start with a certain skill, but your actual skill is HIGHER. Then, you are matched with players below your actual skill and THEN you "outperform". But as the system's assumption about your skill gets closer to your actual skill, your games are a coin flip. EVERY. SINGLE. TIME. At least, if the system works as intended and a sufficient number of players with your skill are available to play...

So, either the system does not work, and it is unable to understand your true skill, or find sufficient players at your correct skill level, which means unfair games. Or the system does work, and every game outcome is random. And, as a side note, "improvement over time" is extremely difficult to influence your ranking, unless you make ridiculously large steps in your improvement (because you are only one of 10 in the game) - it is MUCH more likely (for the largest group of players) to randomly achieve a winrate that is 0.5 - 2% higher over many games than it should be (50%)... and that group will climb to the top of the "normal skill bracket" over a certain number of games. Of course, if you are the top 2% or so of all players, you run through the ranking, but that doesn't change anything for the vast majority in the middle...

The only influence you have is, as you said in your last four words, to be more consistent. But the effect of that on your actual win rate is not as big as people think, as I said elsewhere.

And I already know, you don't believe it - fine with me! If you want to check, do it - the probabilistic model is not that difficult to program if you have basic knowledge of any programming language and some undergraduate level statistics knowledge... just saying, and btw no offense, almost everybody believes that same myth, so it's not just you :)

Smitty Baccal1 votes

if your saying theres no difference between a gold and a silver player, i totally agree with you therefore wether ur 1 or 2 division higher lower doesnt mean ur worse or better than the popele around u. But there are big <<skill gaps>> between the elo tiers. Im currently d2 and having went trough the divisions (all of them) this season i can defenatly say that gold has different isues than plat and plat have different issues than d5 and high diamond is very different from masters. One could probably play a few good games in higher elo than their own.

as for the coinflip part, you are defenatly right if you meant that a bronze 5 game placed against bronze 2 game is still a coin flip. I have found defenatly harder to play with golds than it was to play with d5 but you gotta play the right position and the right champions that can have bigger impacts with the problems your elo is facing in terms of mistakes and skills.

Mephisto31 votes

I am saying if the system tries to make two teams "equal in skill", there are only two possible outcomes: (1) the system worked, now it is 50-50. (2) the system failed, now it's something else.

If the system works, every game is a coinflip (50-50 chances), regardless of division, tier etc. If it doesn't, it doesn't, that's all I'm saying.

And great, of course you are d2 - ALL players who, through random chance or because they are actually at the top 2% or so of the player base, will ALWAYS argue in favor of the system, because if the system does work, it means they are good players! That's very human, very understandable and perfectly fine.

We are discussing this in other posts, if you are interested... regardless, it is still a myth and not true that "if you are better, you will win more over time". It may be true for d2 / top 2% players, but it is not for the majority of players.

Smitty Baccal1 votes

the system is working, it it cnnot take into consideration <<improving>> thats where it gets fun! you are actually placed with poeple of your <<skill level>> from the game right before, if you actually learned and improved, you are now over that <<skill level>>> wich means u have more chances to win

Mephisto31 votes

Well, evidence for why you think it is working is missing... but since you can't have any, we let that slip.

As per the other statement: of course it can take that into consideration. That is part of the whole IDEA of the Elo system... as you improve, your points increase. The question is how it's modeled, again, we don't know.

Karunamon1 votes

regardless, it is still a myth and not true that "if you are better, you will win more over time". It may be true for d2 / top 2% players, but it is not for the majority of players.

How do you figure that? There are two possible things that explain people not moving:

  1. The system is broken
  2. Most people suck at League

I tend to lean more towards #2, but it boggles my mind that personal stats don't count for anything.

Mephisto31 votes

Actually, both your statements are not (per se) correct:

  1. the system could be designed for something else than "climbing to skill", as I have explained (e.g., climbing consistently over time to maximize play time) - so it could very well work

  2. if MOST people sucked, i.e. were of a similar "average" skill level, that just supports what I am saying: you need extreme skill differences to make an impact, the vast majority will simply randomly win and lose (along the various dimensions etc. I have outlined - read my post first)... if you want to call the average player "sucker", sure, suit yourself... I am certain that LeBron believes everybody else but one or two humans "suck" at basketball...

Mephisto31 votes

Oh and to answer the OP question: it is NOT. That is the ONE thing Riot does disclose about how the system works - ONLY wins / losses are considered.

More from Gameplay

3 Refunds per season, upvote if you want 3 refunds per season.
Gameplay764 comments
7,167
votes
3 Refunds per season, upvote if you want 3 refunds per season.
Gameplay156 comments
3,510
votes
UPVOTE FOR URF
Gameplay606 comments
2,440
votes
Yes plz.
Gameplay456 comments
1,721
votes
3 Refunds per season, up vote if you want 3 refunds per season.
Gameplay114 comments
1,716
votes
When is ANYONE at riot going to address the awful new player experience?
Gameplay634 comments
1,383
votes

More by Karunamon

Stop calling every poor call or stupid death "inting"
Gameplay138 comments
127
votes
On 'thin skins' and 'SJWs' and other tired, toxic player talking points
Player Behavior141 comments
85
votes
Meanwhile, over in DOTA land:
Player Behavior98 comments
66
votes
A quick visual guide to the reporting system
Player Behavior44 comments
56
votes
(Rant) Riot, your system makes it very hard to maintain principles of good sportsmanship
Player Behavior87 comments
52
votes
If I were God Emperor of Riot and could do whatever I wanted for season 10..
Gameplay85 comments
49
votes