You must be logged in to post messages.
Please login or register

General Discussions
Moderated by Maffia, LordKivlov, JimXIX

Hop to:    
Welcome! You are not logged in. Please Login or Register.8 replies
Age of Empires III Heaven » Forums » General Discussions » If There Was Ever A Time To Implement ...
Topic Subject:If There Was Ever A Time To Implement ...
posted 01-26-07 10:44 PM EDT (US)         
A variable within the ELO rating system that takes the percentage of win/loss from agecommunity stats, like PR does, it is now. Today just through QS, I have lamed (won every game), been lamed (lost every game), and played very few "non lame/dutch" games. As I see it, by implementing the win/loss stats from agecommunity into the ELO system, wouldn't we have a more accurate rating?

Gameranger: _NiGhThAwK_
posted 01-27-07 06:51 AM EDT (US)     1 / 8       
win prc is very inaccurate. Win prc of top players would be more accurate, but still not great.

To be more precise you should check how often a player wins from someone with higher elo, and how often players lose from players with lower elo. Then you should combine that with the civs that were played.

Or make elo rating for every civ someone plays (so 11 eloratings per player), then it should take the average of all (top) players, and then you can use that.

Doubt they will do that

"such a kind fellow!" ~ ķįŋğ_Ćħŗĩџ_ĬĬ

Furby killer should be crowned leader of AOE forum ~ [SW_GD]Teutonic

James Lock
posted 01-27-07 02:50 PM EDT (US)     2 / 8       
If only you could filter by ELO rating.

The choices could be...

Within 50 cuetech points
Within 100 cuetech points
Within 150 cuetech points
Within 200 cuetech points
Any Skill Level

Thanks to all those that signed the petition to get me unbanned here.
And special thanks to smashnbash for making it.
posted 01-27-07 05:07 PM EDT (US)     3 / 8       
I don't think it's very innacurate, tell me one civ win % that isn't accurate with the real OP or UPness.

If they were to base the rating factor based on civs, then it should at least be based on top 500, if not top 1000. Anything "less" wouldn't provide enough precision IMO.

Gameranger: _NiGhThAwK_

[This message has been edited by Brtnboarder495 (edited 01-27-2007 @ 05:08 PM).]

posted 01-27-07 07:06 PM EDT (US)     4 / 8       
Russia's win % is high, however they are not even close to being strong.
posted 01-27-07 07:39 PM EDT (US)     5 / 8       
I said it many times before, but anyway:
We will not make ELO ratings dependant on the civs win-percentages. Never. And the approach is so obviously flawed that i refuse to believe PR uses civ-win-percentages directly. (Hopefully, the process to determine civ balances/handicaps is much more dedicated, and the whole civ-win% thing is just a myth resulting from poor communication.)

Anyway, we are looking into ways to adjust for civ imbalances on ELO, but the calculation will definitely not be based directly on overall civ win-percentage.

posted 01-27-07 10:34 PM EDT (US)     6 / 8       
In theory there would be no "civ variable" necessary, because the civs would be balanced and have relatively equal win percentages. But in an age of laming and severly OP civs, I see this as one of the only ways to counter act ES' lack of action to make things somewhat fair.

But I'm sure you have your reasons.

Gameranger: _NiGhThAwK_
posted 01-27-07 11:29 PM EDT (US)     7 / 8       
ES just needs to make iro/dutch/spain give absolutely NO PR gain unless playing against another of thse 3 civs.

And you shouldn't lose PR by losing to these civs either.

So lame...

Beren Erchamion
posted 01-28-07 00:41 AM EDT (US)     8 / 8       
or get a bonus for beating OP civs.

IMO not gonna happen, and should not happen - if you did this

it would be like accepting the OP civs for how they are. I

think it would be much better to apply more balance changes,

quicker and more often, then tinkering with the rating system.

You must be logged in to post messages.
Please login or register

Hop to:    

Age of Empires III Heaven | HeavenGames