Welcome to the Forum Archive!

Years of conversation fill a ton of digital pages, and we've kept all of it accessible to browse or copy over. Whether you're looking for reveal articles for older champions, or the first time that Rammus rolled into an "OK" thread, or anything in between, you can find it here. When you're finished, check out the boards to join in the latest League of Legends discussions.


New ELO point system

Comment below rating threshold, click here to show it.


Senior Member


You totally took in to account my team with this data you looked up this on lolking... Or how about asking for taric as vayne and getting nunu (who does terrible with vayne, at least the 2 i got)

You missed my point entirely.

You think that the current Elo system values wins too heavily. You think that because the only thing that matters is wins and losses, that you are personally losing Elo despite being supposedly better than players in your bracket. You have to then also be of the opinion that there are people undeservedly gaining Elo despite being supposedly worse than others in their bracket, because you seem to think you're always better than your enemies. Therefore, any time you lose to a team that you believe you're a better player then, you're losing Elo (despite being better than the enemy team [and in your opinion, probably your teammates also]) and the enemy team is gaining Elo (despite being worse than you). This is YOUR perception of the current Elo system.

And yet I point out to you that your wins and losses exactly match your positive and negative KDA. How do you respond? With blaming your teammates for your KDA.

Okay, you're still going to settle into this fantasy world of "my teammates are always bad, my enemies are always better than my teammates." Okay. Okay, okay, okay.

But your Elo solution to this is some sort of tiered Elo system that will put some weight on things that have little to do with winning matches (and heavily discourages people from playing supports and tanks). So let's really look at your "solution" applied to your matches.

Match 1 in your history I posted before (when they were all ranked matches):
Graves, 9/6/17, victory.

You get +9 for Victory.
+9 * (0.65) for kills = +5.85
-6 * (0.80) for deaths = -4.80
+17 * (0.40) for assists = +6.80
I assume you didn't DC, and I know you didn't surrender.
I don't know how many towers your team downed, but in my opinion, you should also be losing Elo for losing towers, otherwise the tower game is just +Elo for both teammates with no incentive to defend towers really. Best case scenario though, your team downed all 11 towers, so +2.2.
And I don't know how many times your team downed Baron, let's guess once, so +0.5.

Right, so all total that's 19.55 Elo gained, but you also have at cap at 16 points.

In this calculation, you've actually reached your 16 point cap before we count towers or Baron. And given that Towers and Baron are team objectives and Elo points that are awarded to the entire team, it's a pretty weak thing to award Elo for if you want Elo to be based on individual skill. Heck, awarding points for towers and Baron just means people like you will be bickering that you always get teams that can't down towers or Baron.

But let's also calculate your Elo for one of your losses.

Let's take your 7/7/6 with Graves. This is your best performance out of your losses.

So again:
-9 for the loss.
+7 * (0.65) kills = +4.55
-7 * (0.80) deaths = -5.60
+6 * (0.40) assists = +2.40

If we assume no towers killed, no Baron killed, no surrender, no disconnect, then you'd be losing 7.65 Elo for this match. Normally, you'd lose 10-14 Elo. You've shave 25-50% off of the Elo penalty for a very lackluster 7/7/6 match.

How about the match that would cost you the most Elo loss? 3/7/1 Vayne.

Once more time:
-9 for the loss
+3 * (0.65) kills = +1.95
-7 * (0.80) deaths = -5.60
+1 * (0.40) assists = +0.40

With no surrender, disconnect, tower, Baron, you've scored -12.25 Elo for this match. And this is right smack dab in the middle of your average Elo gain/loss. And this score is only made better by your team killing towers for you.

And guess what, a 3/7/1 performance by your AD Carry is basically a no-show. They might as well be AFK. It's hard to win a match with a 3/7/1 AD Carry. I'm not trying to tell you that you played bad, but a team can't expect to win with a 3/7/1 AD Carry typically.

So all that you've really done with your system is figured up a way for you personally to gain more Elo for your wins and lose less for your losses. Yet, if we look at the score board, your team tends to win when you have more kills than deaths, and your team tends to lose when you have more deaths than kills, so Elo was already working for you. All your system would end up doing is inflating everyone's Elo. The only people who are going to be scoring -16 Elo are people who surrender (and I'd be unfairly punished if my 4 teammates agreed and I didn't), people who disconnect (again, cases where people would be unfairly punished for things that aren't their fault), and people who intentionally feed.

The Summoner's Code ALREADY deals with people who make a habit of disconnecting and intentionally feeding.

Comment below rating threshold, click here to show it.

Niiko The Goat



I don't think this system works, even though I'm not level 30 and haven't played ranked games, I can say that I've won many games where my team got less kills and/or assists than them, so that means that killing, dying or assisting is not an indicator about how well you did. Yesterday, I had an enemy Kog'maw, the ******* had above 20 kills at around 20 minutes while none of us had not even 10, he would kill any of us in like 3 or 4 hits (I was Mundo and not even my ult could save me), still we won, they didn't get any of our base turrets, the rest of them weren't bad, we just played as a team, pushed, pushed and won, didn't even need a baron, that begs the question...Did that Kog really played that well??? Did he really deserve a reward??? Maybe not because he was never close to win the game, which is the whole point.

Also, I don't agree with the whole turret thing, even though you have to take down turrets if you want to win, you don't need to take them all down, you perfectly could just make your way through one lane into their base and win and leave the rest of the turrets to rot with full health.

Even though in your proposal you consider having a cap, let's say you're ready to win the game but one or more of your teammates haven't reached the cap yet, what would stop them to forget about winning for a while and wait to kill more enemies or go take down the remaining turrets so the whole team gets max elo points, and yes, you may be risking the win, you may stupidly lose or you can win anyway, in any case, you're lengthening the game unnecessarily because you didn't do what you were supposed to do which is just go and win it.

The thing is, is not the actual system that is flawed, is the people, whining, trolling, being as5holes... and there's not much Riot can do about that.

Comment below rating threshold, click here to show it.




I like the idea however I still think the support is at a disadvantage if on the losing team because they will not get as many assists because there team arnt getting as many kills and still sacrificing themselves for others, also the build tankyer if your getting insta gibed as a support on the losing team is difficult to do because you are still getting very little gold, having to buy wards, and hopefully helping the team with auras makeing additional health items harder to get. To make support less of a elo handycap I would sugest a small amount of elo for healing and or shields, and to help those initiators on the lossing team (not nessicarally there fault for not having a team who can back them up when they initiate) I think a little bit of elo for damage taken could help even that out. However with both suggestions I think it would be a hard amount to get right so that people wouldnt abuse it buy still help those for whom it is intended
just my 2 cents