AOE-II tg elo

[Lokalo]

Isn't it a little unfair to players who have worked hard to just through no fault of their own take away 1000 elo or so in their tg elo?

Hell no

Some of the community is stackers sure, but to penalize the whole ladder because some players don't play many 1-1s is totally not cool and might make some people want to play aoe less and the new system isn't a totally accurate representation.

How 1v1 has anything to do with this?

Although as there is not accurate team game elo system and there won't be if you want to work your ass on elo to brag about it, it must be 1v1, TG elo doesn't mean much.

And ppl wasting tons of time to inflate their elo to 4.5k just by abusing the system are either complete idiots either kids who think someone gives a fk about their 4.5k team rating besides them.

 

[Memeluke]

>tg elo doesn't matter, it's meaningless, nobody cares
>AAAAAAA tg players have higher tg elo than I have 1v1, that must change!!!!!!

pick one guys

 

[AttilaTheHun]

Elo is just a number used in a match making algorithm. Who really cares about the number itself? The change didnt adjust your ranking. If you had the highest rating before the change, then you will still have the highest ranking. Only thing i want is a good working matching algorithm so i got matched against equally skilled players. I really dont care if my rating is 100, 1000 or 10000.

Only thing i am worried about is if these changes (normalizing the TG ladder and adjusting the calculations) are the right think to make the match ups better. I feel like the new calculations are still wrong. So TGs will still be unbalanced.

 

[Memeluke]

it is wrong in the sense that an individual mmr for team games is an unsolvable problem, but it kinda adresses many issues we've had before: it doesn't inflate like tg elo version 1, it isn't abusable by smurfing like tg elo version 2, top players' elo is now lower so you can climb faster if you're good and always win (just haven't played many tg before), you kinda have to play and win more than 50% of the time against other teams of the same rating if you want to climb unlike both versions...

The only obvious problem I see is the players boosted by smurf friends are still high elo, and unless they'll allow it to decay/lose games (yeah, sure...) they'll stay overranked forever (hey at least they don't climb endlessly like before...). FE really should have nuked them, those "negative elo ally" clowns at the top are devaluating the ladder.

 

[Zbyszek]

I have 17++ 1v1 elo and i would have played against 1300s with my 2k4 TG elo, gaining like 4 points per game, and being subject to decay when i am inactive 11111. How can you criticise the removal of elo.

I pointed out on the day that they made the first elo change that they had to take points out of the system. I dont know why they take so long for all these simple changes but this is definetely a step forward. Correct me if im wrong but the elo calculation is just the same as vooblys now.

Anyway... I was matched with only 1900s (TG) when I am 1500TG now and fuill picked civs ofc so it seems to be pretty bad still with regards to matchmaking and such. 1111

 

[AttilaTheHun]

it doesn't inflate like tg elo version 1

I have bad news for you: The new calculation isnt zero sum anymore and thus results in inflation. It is probably not as bad as in version 1. But claiming that their is no inflation on the new TG calculation is wrong.

I just posted this in another thread. I might as well post it here as well:

No, it is again a non zero sum game. In the old bugged calculation it also wasnt a non zero sum game. Then they fixed that and it was a zero sum game. But how they changed it some days ago it isnt a zero sum game anymore.

Just do the calculations for a 2500+500+500+500 team vs a 1000 + 1000 + 1000 +1000 team.
If team 1 wins, then the 2500 player almost got nothing since 2500 v 1000 is an easy win for the 2500 player. A 500 winning against a 1000 has a really small change, so they all get almost the full 32 points each. So that side is getting about 90 points. On the other side it is 1000 v 1000 for all 4 players. That is a 50-50, so they all losse 16 points, which is 64 points in total. So this game would generate about +30 elo and it clearly not a zero sum game.

I was too lazy to really calculate all the numbers, so i used some hand waving for the numbers. Feel free to do the calculations if you dont trust the hand waving.

 

[Memeluke]

I have bad news for you: The new calculation isnt zero sum anymore and thus results in inflation. It is probably not as bad as in version 1. But claiming that their is no inflation on the new TG calculation is wrong.

You forgot your elo decays as well if you don't play for 14 days or disappears if you abandon your smurf forever... 1v1 elo is not zero sum as well, elo disappears and gets created (like when a tg player does his first 1v1 and goes 2k instantly) but it's fine enough imho, it won't inflate nearly as much as tg elo version 1 which used highest elo instead of average for calculations (so every single game ever created more points than it's gonna do now).

 

[komtan]

change rate and cry
we love this classic (^_-)-☆