Skip to content

Possible solution for "elo hell"

TheVeryVeryBest··6 votes·815 views

What do you guys think, using the new champion mastery in ranked provisionals to see who had the most ward coverage or kill participation or turrets destroyed, see if they win or lost, check averages for that champ in certain tiers (I.E. Diamond or Bronze tier ratings) and place them according to there average placement after there 10 games, now trolls can't force you into bronze and it will be much harder to be carried by smurfs or people who are just plain good. I personally think that this would be much more effective and provide a healthier environment for poeple who feel they got queued with "feeders" or "trolls" then the current system! Please share your opinions.

7 Comments

Franco Sanchez3 votes

Except stats have nothing to do with individual skill. Your proposal doesn't include ways to show EFFECTIVE ward placement. It doesn't show whether you are mechanically better than the people you are playing with. It doesn't show that you have better decision making skills. Or knowledge of items and builds.

The only thing stats (K/D) show is that you were better, in that game, than the people you were playing against. Or more lucky. Or more fed. Or they made more costly decisions. Or all of the above. And a warding score is really easy to jack up, when you are sitting in your base, losing, and making sure you have fantastic vision between Nexus and fountain.

Good idea in theory, but loses points in practicality.

jaymc11303 votes

[{quoted}](name=Franco Sanchez,realm=NA,application-id=3ErqAdtq,discussion-id=7T3oAHdp,comment-id=0000,timestamp=2015-05-16T08:25:01.176+0000)

Except stats have nothing to do with individual skill. Your proposal doesn't include ways to show EFFECTIVE ward placement. It doesn't show whether you are mechanically better than the people you are playing with. It doesn't show that you have better decision making skills. Or knowledge of items and builds.

The only thing stats (K/D) show is that you were better, in that game, than the people you were playing against. Or more lucky. Or more fed. Or they made more costly decisions. Or all of the above. And a warding score is really easy to jack up, when you are sitting in your base, losing, and making sure you have fantastic vision between Nexus and fountain.

Good idea in theory, but loses points in practicality.

^ It's a thought in the right direction, certainly. But what you propose isn't very well thought out yet.

There's no reason champion mastery shouldn't affect they way the ranking and matchmaking systems work, and I rather believe we will see some integration here over time, but this is all still prototype stage. It's gonna be a while before we can talk about a production model.

JustMyBassCannon2 votes

So make the system smarter than the troll.

We can build a computer that can beat 1000 people at chess all at the same time. Hell, League's system can already keep track of when and where a champion kill happened, or when a turret/Baron/Dragon was killed, and who was involved in any of those cases. It stops short of tracking each individual minion and/or monster kill.

I think we can build a program that keeps track of ward locations, when it was placed, whether or not that location already has vision, how often it was used as a Teleport node, if it was killed prematurely and by whom, or in the case of Vision wards, if/when and by whom it was killed, etc.

We could probably also build a program that compares each user's personal MMR, so it can scale the score differences between players more fairly. "Yeah, you had a pretty bad game, but you queued with your buddy who's several divisions above yours, and all of your opponents were above yours as well. Your score is still going to be penalized, but not as severely as if you'd scored this on a level playing field."

We might even consider going as far as tracking specific champion abilities. How often did the Brand player chain his Blaze effects? How many axes did that Draven player catch? How many people died after stepping on one of Caitlyn's traps? Did Mordekaiser steal their carry, or their support? How much AP did Veigar farm with his Q?

Now, I'm not naíve. I'm actually a programmer myself, and I am fully aware of just how daunting a task this suggestion is. But the system we have now is just not good enough. It often rewards poor play because the player made enough smart plays to help their team carry them, or punishes great players who got bad teams, picked into bad compositions, etc.

It's an insult I take personally to know that I played 3 Shyvana games to get just barely <200 points more than I got from 1 Taric game, all because the Taric game was a victory and two of the Shyvana games were losses.

Franco Sanchez1 votes

Daunting is a bit of an understatement. And if they found a way too honestly implement a system that adequately measured player skill? I'd be all for it. But there are so many subtle nuances in a game of league of legends, it's rather astonishing. There are so many things that a Diamond player would understand, recognize, and do that are completely different from a Bronze player. And a lot of those things aren't as easily measured as even some of the things that were already mentioned. A huge part of League of Legends skill levels are intuitive, and what really separates the best players from the worst is mechanical skill, decision making, knowledge, etc.

But honestly, the premise of this thread is a little flawed. Elo hell doesn't exist. It's just some construct that players like to hold too when they lose, rather than face the fact that they just suck at the game. And those players that are "stuck there" honestly wouldn't rise, regardless of the system that's in place.

jaymc11301 votes

[{quoted}](name=Franco Sanchez,realm=NA,application-id=3ErqAdtq,discussion-id=7T3oAHdp,comment-id=000000010000,timestamp=2015-05-16T09:10:01.185+0000)

Daunting is a bit of an understatement. And if they found a way too honestly implement a system that adequately measured player skill? I'd be all for it. But there are so many subtle nuances in a game of league of legends, it's rather astonishing. There are so many things that a Diamond player would understand, recognize, and do that are completely different from a Bronze player. And a lot of those things aren't as easily measured as even some of the things that were already mentioned. A huge part of League of Legends skill levels are intuitive, and what really separates the best players from the worst is mechanical skill, decision making, knowledge, etc.

But honestly, the premise of this thread is a little flawed. Elo hell doesn't exist. It's just some construct that players like to hold too when they lose, rather than face the fact that they just suck at the game. And those players that are "stuck there" honestly wouldn't rise, regardless of the system that's in place.

It's no where near as difficult a task as you make it out to seem. The most difficult part, the part that I've been demanding for ages, a way to individually measure each players performance and contribution to the game, has finally just been devised; champion mastery. Now all that needs to be done is tweak the mastery system to better reflect actual contribution vs statistical accumulation and to incorporate it into the matchmaking and ranked systems. The hardest part is done, and that was coming up with a way to quantify individual performance.

TheVeryVeryBest1 votes

What you mentioned last in that statement is the exact problem I see, now I'm not savy with code myself so I would not fully understand the steps do fixing these problems but I myself feel that having a losing and a winning score be different from eachother would help to solve it. Let's say you played Lux support and went 1/4/25 placed wards smartly, shielded your carries at the perfect times in fights and won the game, you receive and A, now if you had lost the game but still done everything in your power to set up kills for your carries and ganks for your jungler then you should also receive and A, I don't expect anything close to this to come for at least a couple of years, maybe a riot employee will make a code which perfectly addresses my concern, maybe not but either way, it would be better then what they have in place now, where a simple fizz who goes 30/2 every game carries an akali who goes 2/10 every game and that player gets silver, they clearly do not deserve silver, so a program which looks at the avarage KDR and stats of a player to pick there ranking would perfectly solve all of riots "elo hell" complaints, I could not begin to know how they would go about this, unfortunately (all I know is sites like lolnexus can show you stats, why can't league use something like that?) your input is strong, you clearly understand more then I do.

ZeeDrakon1 votes

The porblem here is that stats dont mean much. You can be 50/0 and when you lose its pretty much your fault because you didnt do anything with it. Wards are completely iseless since you can sit in base and spam them, turrets destroyed are equally stupid for obvious reasons, and your stats only show how well youve dont against your enemies. You obviously can have better stats in silver than a diamond player in diamond. But thwt doesnt mean your better than silver

More from Gameplay

More by TheVeryVeryBest