Solving the [Solo] Rank Conundrum

Comment below rating threshold, click here to show it.

Elsabet

Senior Member

02-04-2013

The problem- Solo/Duo is not an ideal measure of individual player skill. Players sometimes suffer terrible streaks of misfortune in a less-than-perfectly-balanced matchmaking system. Rubber-banding is frequent, as is feeling trapped where one "does not belong." Whether you believe LP-hell exists or not, there are certainly players unjustly suffering from being unable to carry their team of 4, or their team of apparently completely incapable players, or both. Sometimes it feels like the Bronze IV - Silver V range is just filled with decent to horrible players, all just being randomly paired and infinitely rubber-banding. Anyway.

Everyone knows it wouldn't be balanced or fair to award LP based on KDA, CS, or any other automatically measurable factor. It is similarly unfeasible to expect Riot employees or the player community to watch ranked matches- especially low level ones- to selectively award certain players who had the misfortune of playing with bad teammates.

So how do we get players to the rank they belong at, so that they can play more balanced, competitive, and rewarding matches? Simple: let their play tell the story, and let it be determined by those in the best position to judge- those they play with.

In the post-match summary, each player is given the opportunity to rank each of his teammates and each of his opponents, or choose to abstain from assigning ranks in the case that he was unsure- if, for instance, he did not see compelling reason to rank one person's play over another. A player can also give tied ratings, for instance rank an opposing Ahri as 1, and give the opposing Cho and Jax 2nd place, leaving 4th and 5th place unassigned.

A player on the losing team with a strong majority of 1st place votes by his teammates and opponents would gain more LP than a low ranked player on the winning team. No players on the winning team would lose LP. (Actually I'd be open to someone with a unanimously- among allies and opponents- low rating on the winning team losing rating. But that's a separate discussion.)

Another way to do player rankings, as opposed to 1-10 or 1-5 for each team would be for each player to have a limited number of designations to assign to their teammates and opponents. For instance, I can assign one player (aside from myself) as having "Played above his/her rank," and another as "Played below his/her rank." Or to simplify it further- "Played well" or for a player who played very well on a losing team, "Does not deserve to lose rank for this match." After all, that's what we're all concerned with. Yet we all see it happen to our opponents all the time and we just say, "Welp, glad that wasn't me!" This sentiment has no place regarding incidents on the Fields of Justice.

The main goal of this institution is to protect good players from losing rank due to the play of their teammates. Players won't always vote accurately, but LP gain/loss will be based on the popular vote. Let's face it- most matches, players recognize and sympathize with the best player's performance on the losing team. Under this system, you'd still gain the most rank by consistently winning, but players who consistently either win and/or play well, will see their rank consistently rise. This would be a good indication of individual skill (and the higher the rank, the more accurate), worthy of being labeled as a "Solo/Duo Rank."

Players would be discouraged from paying inordinate attention to KDA and to try to base ratings of players on their apparent, demonstrated skill. Players would similarly be discouraged from assigning these skill ratings based on player personality- that is what reports and honors are for. This unrepresentative rating can't always be avoided, but again, we're talking about a general measure of improvement for the current system.

Some possible remedies to unwarranted ratings/favoritism:
1) Players do not rate their duo queue partner, or themselves of course.
2) Perhaps players only rate their opponents, or their teammate ratings are completely separate from their opponent ratings.
3) Players' rating histories are automatically compiled and a tribunal case is generated if they too often rate skill in stark contrast to their peers. A possible punishment would be losing the right to assign ratings.

I think players would be grateful for the opportunity to play a part in the rating process of their peers, as opposed to helplessly watching as good players are dragged down and "noobs" get carried undeservedly. I believe this system would be quick, unburdensome, fun, and useful.

Thoughts? Contributions? Any reason to fear that this would not at least somewhat improve the accuracy of the current "solo" rank designation?

Thanks for reading,
Elsabet


Edit: Please read this post, not responses, to understand what I'm suggesting.


Comment below rating threshold, click here to show it.

CoOokiesX

Junior Member

02-04-2013

I understand your point but this would only be an opportunity for the trolls to troll more.

If Riot did the judgement based on individual performance then didnt deduct points, ranking placement, etc for only the first place player on the losing team, and then a decay to the 3rd best player so the losses wouldnt hurt as much I think people would be motivated to peform better even when they have trolls and afkers.

Winning a 3 on 5 or 4 on 5 game would give a larger reward, a promotion to the next division and or bonus points, etc. This would help keep the team working and fighting hard because something will come of it even if it does end up being a loss.

Removing the power that trolls have has to be done at some point and Riot has had the tools and data to correct this for years. Maybe some day.


Comment below rating threshold, click here to show it.

adc

Member

02-04-2013

How would you determine who the best player is?


Comment below rating threshold, click here to show it.

ROKAF JE

Senior Member

02-04-2013

It's not measuring individual skill really, but all players within a certain bracket.

If you play consistently, eventually you'll be lumped with a group at your skill level.


Comment below rating threshold, click here to show it.

ROKAF JE

Senior Member

02-04-2013

Quote:
Originally Posted by AD Bottom View Post
How would you determine who the best player is?

Stats? Attitude?


Comment below rating threshold, click here to show it.

adc

Member

02-04-2013

Quote:
Originally Posted by ROKAF JE View Post
Stats? Attitude?
Stats don't mean anything - I've been carried and gone 9/3/7 compared to the rest of my team going negative. And attitude isn't an indicator of skill at all, so basing the rating on that would be stupid.


Comment below rating threshold, click here to show it.

HEX Axis

Member

02-04-2013

Bump, I think with a few changes made by Riot staff this could really help the elo system, and with this new rating system just being put in, what better time to do it?


Comment below rating threshold, click here to show it.

Orcman21

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

Recruiter

02-04-2013

Why should I reward somebody else in a match? There is no incentive for other players to up-vote their teammates/opponents. I would see it as just another button to click at the end of match like Thumbs Up and Reporting. Even now if I was to play a solid game there is the Teamwork Ribbon which I'll get. No need to double dip on rewards for playing good.


Comment below rating threshold, click here to show it.

HEX Axis

Member

02-04-2013

bump


Comment below rating threshold, click here to show it.

HEX Axis

Member

02-04-2013

Quote:
Originally Posted by AD Bottom View Post
Stats don't mean anything - I've been carried and gone 9/3/7 compared to the rest of my team going negative. And attitude isn't an indicator of skill at all, so basing the rating on that would be stupid.
You rate according to your own opinion who the best player on the enemy team was. There's no wrong answer, if you noticed that a certain individual made a lot of plays happen, but went 0/9/7 you could still rate them as 1.