Ranked Ladder?!

Hey Uber seriously?
This ladder is a joke. If someone is UBER it means NOTHING, because if this:
Right now 26.11.2014 (16:15 GMT+1)

UBER: 48
PLAT: 31
GOLD: 15
SILVER: 13
BRONZE: 8



41% of your ranked playerbase is UBER

Shouldn't be the system make sure that only the best of the best are in UBER? So that most of the players are ranked lower like in LoL?



The system now is like in the kindergarden, where they only play games in which all can win.

tl;dr: being UBER have to be something special

I know, it doesn't make sense. Let's wait for a response from Ub- oh wait.

The ladder is less than a day old. It's barely even started being populated. If your graph still looks like that after 1-2 months then by all means come back and complain about it. But this is extremely premature.

It is like that, because a huge percent of the would-be bronzes aren't ranked at all. If they were, they would outnumber the Uber.

That being said, a few low percent can feed the few high percent 5 wins, and the rank is based solely off ratio and not weighted by how good your opponent was. If you win 5, even 5 bad games and the 5 never place, then you are Uber regardless.

To change this, they would need to grade people based off who they beat and that player's rank. Thus, to be an Uber, you need to beat many platinums or a few other Ubers, you can't just beat a sheer mass volume of bronzes...

Also...

Since we dont know what ladder algorithm is used we cant tell if this will get better.

The PTE ladder didn't seem to get better.

I don't think there are enough people playing the PTE to say whether it got better or not. You stomped me in one of my 5 ranking matches and I ended up in bronze... but after playing 2 more matches I moved up to Silver. I think once we get more people playing the ladder it will even out, but it's hard to judge from just the PTE sampling.

Currently they seem to be going for an even distribution in each league. Instead it should probably be more like [ 25% - 25, 25% - 25, 25% - 25, 25% - 25, 100]. This would leave every league even except the Uber league, with just 100 players at any time.

PTE and stable were using different matchmaking queues but the same ladder.

Yes but IIRC the old stable ladder had the same issue.

You might find this an interesting read:

http://joostdevblog.blogspot.co.uk/2014/11/why-good-matchmaking-requires-enormous.html

this makes complete sense. Think about it: What kind of player is most likely to be ranked less than 24h after a reset? The casual bronze player? The competitive uber player?
It makes perfect sense that Uber players are far more active and are far more likely to complete 5 games to be put into a league within 24h.

The numbers they gave us for PTE suggested not, rather the largest population should be in silver. The issue is people in placement are ranked, but don't show in the stats, though they do count for distribution.

The old ladder also had this problem. About a third of folks were UBER.

IIRC, each of the tiers in the old ladder had roughly an equal number of players in it.

~280 for Uber and ~170 -> ~220 for the rest if I remember correctly.

It was an equal number of visible players, not an equal number of players.

I don't mean the fact that only the top 10 were shown . I just recall checking the counts once or twice and they were roughly the same. I'm not active on the ladder though, so take other peoples' word over mine .

This was clarified at the time that individuals who hadn't completed placement were contributing to league distribution. There were a lot more players in silver than were shown, but they didn't complete placement and so weren't included in the count.

Yeah assuming this is using some kind of basic forecasting model, the number of players is just too low right now for the system to stablize. It's going to take some time, and probably some serious user growth before this works right. Just pure speculation.