Boards›Gameplay

What I believe the ranking system should function like.

Hvsao·5/30/2019, 7:14:21 PM·2 votes·1,069 views

Hello everyone, you can call me Joe. This is my second account. I got hard stuck in my old account to a point where I couldn't find the game fun, so I wanted to start again from the ground up.

The current impression I've received from league's rank system has been rather poor. I don't think the system truly demonstrates a players capabilities, and often leads to poor skill development. This issue is primarily with the solo/duo queue ranked matches.

As a solo player I expect that my growth as a player, and my contribution to the team should be the primary factors that determines my rank. I do not believe that I should play at a higher elo if I did not positively contribute to my teams chance at victory, and I believe my teammates should be held to the same standard.

Now the question you should be asking is, "How will you determine if a player has actively contributed to the game?"

-Gather the general statistics of player performance in each division. Generate an average with a small deviation which will act as the baseline for performance metric for each role.

Example:

Jungler: Does the jungler at least meet the average jungle camps cleared for X division? -2 / -1 / 0 / +1 / + 2 Does the jungler at least meet the average KDA for the role for X division ? -2 / -1 / 0 / +1 / + 2 Does the jungler at least maintain average vision control for X division? -2 / -1 / 0 / +1 / + 2 Does the jungler at least capture average number primary jungle objectives for X division? -2 / -1 / 0 / +1 / +2 Does the player maintain a win streak? +0 / +1 / +2 / +3 / +4 / +5 Is the player reputable (honor 2 and above)? +0 / +1 / +2 Does the player win the game? -10 / +10

Exceptional Performance (winning the game) will net you: 18 to 25 LP Exceptional Performance (losing the game) will net you: -2 to 0 LP Good Performance (winning the game) will net you: 14 to 21 LP Good Performance (losing the game) will net you: -6 to -4 LP Average Performance (winning the game) will net you: 10 to 17 LP Average Performance (losing the game) will net you: -10 to -8 LP Poor Performance (winning the game) will net you: 6 to 13 LP Poor Performance (losing the game) will net you: -14 to -12 LP Horrible Performance (winning the game) will net you: 2 to 9 LP Horrible Performance (losing the game) will net you: -18 to -16 LP

Your performance will directly influence the LP you gain. There is no hidden metric which will work behind your rank to determine your LP. The higher the rank, the more the game will expect from you. Playing poorly will only reflect poorly on you. This system will always provide more LP per win then Loss as long as you are consistent.

Trolls & Afk's will not benefit from their actions, because its effect on you will reduced.

The games overall outcome (team oriented) accounts for 10 - 15 LP While individual performance and attitude accounts for 10.

I believe this is a fair balance which will provide players with opportunity for growth in solo/duo queue.

11 Comments

NoPaxt5/30/2019, 7:23:11 PM4 votes

Here's what the ranking system should look like.

Did you destroy the Nexus? Yes +1 or No -1

Slack King5/30/2019, 7:20:53 PM3 votes

Does the jungler have a certain camp clear score Well guess shaco gets massive lp losses Does he meet the kda Well I guess tanks are out Vision control is a team thing jungle objectives can be massively a team objective

All these things arent good, they cause people to worry more about stats over objectives Which is probably why youre stuck

Saezio5/30/2019, 8:11:52 PM2 votes

Straight up no. First of all, you would need to have different values for every different champion because different champions need to focus different things to be played optimally. Farming vs Ganking junglers for example. Secondly and most important, people would not care about winning, just about stats. People wouldn't end to make their scores better. And finally, it's a team game, you win or lose as a team.

Kai Guy5/30/2019, 9:45:44 PM1 votes

This is a common enough suggestion, that gets shot down frequently. I like that you tried to break down and create some actual numerical values. Good on you to make effort in your posting.

IMO the largest issue is that the stat system always has to be subjective. It needs to be able to cast judgment on if stats are significant or not on a match to match basis or its open to abuse and will collapse when its proven abuse is the best way to climb.

Saezio hits several valid points. Champion stats need to be tracked individually because champion kits are not equal. Not every champion has the same potential for damage. Even in similar classes champions vary. MasterYi is unlikely to ever get to high of a CC score.

I don't think you can trust the league community to not attempt to exploit stats systems. You already can see a high number of players who are extremely selfish in the play patterns to protect KDA currently with no reward. Historically in games (Blizzard ) I played with stats added to MMR systems I never saw it create any better quality MM, but it did change some playstyles, anecdotal so take it with a grain of salt.

The Positional Que system was ambitions but not entirely awfully on paper. Targets a lot of player pain points with bad teammates, OTP, Auto fill. It had to be scrapped because they player base abused it (low MMR role swap to main) and frankly cared less about their off roles then they shoulda because it did not impact their mains if they did so. Made for less competitive matches not more.

The value of stats is always very subjective as well. Like I said. Padding is easy. Karthus who hits R every time its off CD and there are 5 enemys alive will generate more damage stats from that skill then one who saves it for enemys who almost got away or if a fight breaks out and decided to help his team.

Saving ability's for the right time is a part of skill expression in this game. You don't want to Burn your CC vs Katarina. You need it to cancel her R. But if you spam your CC you get more damage and more CC score Stats. So your stat system needs to be able to tell the difference between meaningful or padded stats. And padded stats raise the expectation for the average you have an issue with the more people who pad results in less accurate skill for people playing to win.

Important to point out. KDA always makes for selfish players. the impact earlier Deaths is huge. Lets use 20 for easy math. 10-0-10. Flawless. Then lets do count up to 5 deaths starting at 1.
20 to 1, 10 to 1, 6 to 1. 5 to 1 4 to 1.

Lets say it was a rough game that had a lot of really close team fights. To preserve your KDA you want to run when your at risk. If you don't, and you died a lot but for the right reasons. 12 deaths makes your kda 1.6 to 1.

A stat padder will avoid the D because they know that it makes dramatic impact. You cant really separate it... Make it so Deaths as a stat don't matter? Seems unwise.

In league you can die for a good play or bad play. Your system needs to know which one it is. Same for CC, for clears, for objs, for damage and vision and CC. This makes it subjective, it has to be able to look at the game. Subjective, automated systems are hard to build.

The most common way is to tack them onto MMR/Elo systems. They just off set how much you gain/lose. This does not reward all players equally if its flat stats. Its extra work to build for each champion and can make negative impact for less popular picks with dedicated mains. Plus with that solution end of the day your still using MMR so you will find players getting stuck and frustrated.Being stucks not fun but MMR allows for you to climb or sink always.

I am a outspoken fan of Elo systems and spent a lot of time learning about them. If you have questions specific to Elo/MMR feel free to bug me.

ZephyrDrake5/30/2019, 10:33:43 PM1 votes

When will people realize that playing for stats is easier than playing to win?