Ranked ELO and some positve changes that could be made to it.
ELO Hell that's what it's been called, and many will point out to you that it is just a concept created by players who simply don't know how to play the game or are unable to use team work. Let us not discuss the personal merits of one individual but the over arching concepts which make ranking a team based game so difficult. In fact I will not discuss ELO Hell at all as it is inflammatory and polarizing. Instead I will discuss some of the various problems which face a team-game ranking system with a solo or duo Que option.
The ELO system, no matter how it is devised, will always be flawed when applied to individuals, because it is based on algorithms applying specific rules to individuals based on the performance of the entire team. Let me reiterate, You cannot accurately judge the performance of one individual based on the performance of a team. There are many reasons why this is so, but let me just list out a few.
Synergy. A 5 person team has 5 different roles to play generally. Each member of the team picks a role by choice or is forced by circumstance or other team members. It is a scientific fact that everyone is better at certain roles. Certainly there are those that would argue that point so I will rebuff. There is no way for anyone to be just as good at all roles and all champions. It is not scientifically possible because a person can not play all roles all the time all at once and therefore they have had more practice in one thing than another. This is all to say that In a team game with a random sampling of individuals at a relative level of experience it is probable or even likely that a majority of team members will be out of their element or even playing poorly due to the role or champion that they were forced to pick. Furthermore this is even more contrasted when play style is considered especially when considering the early game roles of Bottom lane assuming META. Two players are thrust together in a lane and asked to learn each others play styles in a very short time. Never having played together they have no way to know how they will play, when they will go for kills or how CS is to be taken. They may not have the best champions to complement each other and the may not have play styles which have synergy at all. Many times this is the case. People will argue that a player should have flexibility. I Agree. Which is why I believe that a persons flexibility in roles and with other players should be considered when ELO status is rewarded to them.
AFK. We have all played with people who have gone away from the computer at certain points during the game or simply disconnected. Currently this is supposed to be an offense rather than an element of ELO consideration. Supposedly a person who AFKs will simply be corrected though the tribunal system. However this does not seem apparent as everyone frequently has AFKs on their teams or someone gets upset and decides not to play. I believe that AFK should be considered in ELO in a specific way. It seems that often a person who frequently AFKs loses games for his or her team and that of course will reflect on his or her ELO as if it was just a normal win or loss. So these players are considered by the system to be lower ELO simply because of the loss and placed accordingly with better players to balance out the game. AFK should be considered an individual play style and those who do it should not be picked by the ELO system as a counterweight. This is a problem which has no solution. The game could track statistics and if a person is found to be inactive at a greater rate than the majority that player could possibly be docked ELO rewards to a greater extent. However I believe that no one, no mater how low or high their skill level, should be placed with someone who frequently AFKs. I suggest therefor a special penalty rank that places them exclusively with people who AFK until they prove that they are no longer going to be frequently inactive in games.
Server Lag. This requires little explanation. There are times when everything is well and all players experience lag. It happens. Wins and losses especially in this case are a very poor way to judge an individuals merits. Possibly there could be no points rewarded at all this already happens on occasion but usually when the Lag is so obvious as to be ridiculous. This is supposed to be a highly competitive game an E-sport even so if the server is lagging even a little it can effect the games drastically. I have read an article that RIOT is working on dedicated servers, and that makes me happy.
Social Graces. A few summoners are down right rude and hateful. Others are sensitive to the point of not being competitive because of words. This can lead to wins and losses that might have had a different out come. Yet another reason why simple win loss statistics are inadequate. Fix this? I don't think so. When someone is courteous it can improve the game however if emphasis was focused on how well a person does in a specific role more than win/loss players would find themselves matched with those who are at their skill level of all social stripes. Leave social judgement to the tribunal. Judge a player not by simple wins or losses but by their score and also by their willingness to help others.
Players who focus on Kill/Death scores. You have also played with a such a person or maybe you are one. They tend to focus on having a positive score at the end of the game this leads less team play. These players abandon a fight earlier than others and go back to their respective lane or position. This is one of the biggest problems with an ELO system that would reward individual behavior because people who focus on K/D primarily fail to take part in team play. In a game many players, even with a low K/D score deal out tons of damage to the enemy, tank frequently, place wards or work well with the team to lead to victory. Again this is a play style that has no synergy and can effect the Win Loss. KDA Does not necessarily point to a well skilled player and there for any merit based algorithms should take things into account. Such as whether a person is present to with the team when there are team fights, ward placement, roaming and etc are examples of positive team play.
In conclusion there are many elements that lead to victory or defeat. An individual is one member of a five man team placed against another 5 man team. An individuals performance is only one tenth of the input to a loss or victory. There fore it is impossible to rate a persons skill by wins and losses only. Perhaps a player who does well in a game should get bonus ELO points when winning and take less of a penalty when losing. Perhaps there should be a vote at the end of the game by both teams who was the best player on either team. Perhaps players who frequently AFK should be placed in a special ranking system for a time. What do you think? Should you be judged according to win/loss statistics alone?
