PR is the biggest joke ever so yeah EP ELO is better. Even jap ELO is better than PR anyway. EP ELO is also better than jap ELO because it's played on a patch without cheaters/broken balance/broken maps. But that's just because EP is better than RE; the ELO system itself isn't any better, it's as mediocre as the jap ELO one. But again, we can't really hope for better than that, since there isn't a good way to measure skill.
If you'd seed a tourney right now based on ELO, it would be a huge joke. Then again I guess people would try to get a good ELO to have a good seeding if a tourney was announced, so the rating would be different. We'll have to see next tourney I guess. But for the record, the seeding of previous tourney using EP ELO was bad, probably because it was so new then.
LoOk_tOm wrote:I have something in particular against Kaisar (GERMANY NOOB mercenary LAMME FOREVER) And the other people (noobs) like suck kaiser ... just this ..
EP Elo and an up to date elo in general is a good picture of the current state of players, so it is accurate. It doesn't have to coincide with tourney results or anything, altho if tourney games were rated the two would certainly tend to overlap.
The problem is no quick search. People usually only play against the same players. It's hard to gain and keep gaining any points when you dont have anyone to beat up on, points just keep trading back and forth between the same 20 players.
I think that was a temporary issue for a few hours when they moved the data.
Nah. Simply people were arguing it wasnt accurate. Now Diarouga is in it and magically the ladder is fine. Wow I didn't think I would have to explain the joke.
It's not fine, it's just better than the jap elo now, while it used to be worse.
An ELO ladder gets better when 1) there is more interaction between the different layers of the community. Pr20's need to indirectly be donating their points to the pr40's for the ranking to become accurate. When the interaction between those two layers is only done by a handful of people that play with both the stabilizing goes extremely slowly 2) there needs to be a reasonably big volume of games for the ELO to be accurate. I don't know what the state of the EP elo is like right now, but if the amount of games played is low it's going to be inaccurate
Also, although pr isn't the greatest due to some exploits, one thing that I do think works well is the general volatility of pr. Going from one rank to another takes a bunch of games once you've played enough games (for newer accounts its not great though) to the point where I think under normal circumstances the pacing is quite good. For ELO games, it feels like getting a decent streak can possibly have a huge effect on your ELO. The problem with that is that aoe3 is a volatile game, where you can easily play a bunch of above average games where you also get lucky with the match up/crate start/map spawn and that could really allow you to jump in the ELO rating by a bit too much imo.
momuuu wrote:An ELO ladder gets better when 1) there is more interaction between the different layers of the community. Pr20's need to indirectly be donating their points to the pr40's for the ranking to become accurate. When the interaction between those two layers is only done by a handful of people that play with both the stabilizing goes extremely slowly 2) there needs to be a reasonably big volume of games for the ELO to be accurate. I don't know what the state of the EP elo is like right now, but if the amount of games played is low it's going to be inaccurate
Also, although pr isn't the greatest due to some exploits, one thing that I do think works well is the general volatility of pr. Going from one rank to another takes a bunch of games once you've played enough games (for newer accounts its not great though) to the point where I think under normal circumstances the pacing is quite good. For ELO games, it feels like getting a decent streak can possibly have a huge effect on your ELO. The problem with that is that aoe3 is a volatile game, where you can easily play a bunch of above average games where you also get lucky with the match up/crate start/map spawn and that could really allow you to jump in the ELO rating by a bit too much imo.
Because the playerbase was seemingly slightly larger and thus the top end had higher ranks. This has dropped since the cheaters I feel like, so right now the highest ranked elo players are just the inactive players because actively playing will cause your rank to drop.
momuuu wrote:I'm pretty sure thats good. The jp ELO has actually been deflating. I believe Mitoe just used the incorrect term.
Yes it is good. Just saying that despite the low number of players on EP it is still enough to slowly stretch out the Elo separating the top players from the base rating.
One problem is that since the average new player on ep is between major and capitain, it is the same winning vs a major and winning vs a ms that has never played on ep (which is probably why eaglemut was preseeded master tourney for instance). It means that you can get a colonel elo since bashing major (what a colonel does) is the same like bashing ms (what a lt does).
bwinner1 wrote:One problem is that since the average new player on ep is between major and capitain, it is the same winning vs a major and winning vs a ms that has never played on ep (which is probably why eaglemut was preseeded master tourney for instance). It means that you can get a colonel elo since bashing major (what a colonel does) is the same like bashing ms (what a lt does).
Yea that was the big issue, you could get a high elo without being a top player, while it's impossible on the jap ladder (you can't get really more than +250 elos if you bash noobs). We don't have access to the full EP ladder atm, but it seems that the current top10 is the real top10 although some are overrated/underrated because they don't play atm.
bwinner1 wrote:One problem is that since the average new player on ep is between major and capitain, it is the same winning vs a major and winning vs a ms that has never played on ep (which is probably why eaglemut was preseeded master tourney for instance). It means that you can get a colonel elo since bashing major (what a colonel does) is the same like bashing ms (what a lt does).
Yeah, its a huge downside of small playerbase ELOs. The influx of new players will distort the elo quite a bit too.
Honestly some of these aspects make me appreciate pr.
edeholland 
Kaiserklein 
