Thats for sure the most reason for beeing kicked out of the lobby... more than a bad ping ^^. I like my stats, but I don´t like to look at the stats to others during a game or in the lobby. I only hope, it will be a funny game and no one leaves a running game because of 3vs0 kills or the like.

Greets Tenbian

rankings are dumb. u lose way to much xp from losing to the highest ranked players.

and people who havent played after reset are better ranked than most people.

Stats are useful and important so that auotmatch can match players of equal skill and therefore everyone has a good time.  But of course, Automatch is completely f*cked so yes, there's not much point.

I agree, the stats are pretty much totally meaningless at this point, and ranking especially. I play about 50% pure PUG and 50% with my brother (though most of those are 3v3), so I hardly ever play full-on premade, and as such, I feel pretty good about my slightly >50% win (though I wouldn't call myself an awesome player at all, probably closer to average). I have had a fair amount of PUG games where we beat a PUG stomper group who are all ranked much higher and have an amazing win% when my team was all around the level I was. I think that while there are some awesome players who are top ranked and deserve their win%, but I also think that there is a large pool of pretty skilled players with a 40-60% win rate due to playing a majority of PUG games, and their rankings all suck. Right now, kicking based off of win% or rank is not justified as it is often times not a good reflection of skill... though perhaps kicks based on number of games played is. But even then, the new beta overlay is showing an odd number for games played.

Then you may be excluding perfectly good players who choose to play PUGs more often than not, and have a low %win as a consequence.  Just go by games played.  If they've got a few hundred, they've been around long enough to be experienced and know what they're doing.

Win % is horribly skewed.  Go check one of those top 20 players games-list, where they all have the same game name, and play with the same two people all the time. It isn't evil to play with your friends, but it is interesting to see that most games end up with 3 players who have 70+ win percentage, against a team with 15 to 30% win percentage... except for the few that they lose.

I'm really starting to think there is an interesting pyschological experiment to be had here.

I always look at total games played... I have the naive belief that if you've played the game 700 times you'll generally be better than the guy playing their 50th game.  Most low win % are puggers so the % indicates nothing about their overall skill IMO. 

There are a lot of things they should do like League of Legends.

But the trick here is that ranked games can only be automatched.  That way you gaurantee that people of nearly equal skill are playing eachother, thus giving the win/loss percentage some actual meaning.  This is how it works in almost every game with some sort of ranked stat tracking.  Using custom games to determine rankings is practically unheard of.

But see my previous comment about Automatch in Demigod being totally screwed up.  1) With the insistence of AIs, and 2) The light side/dark side division severely hampering the speed at which matches can be made.

And it's not going to get any better because instead of making a more robust automatch system, the are investing in adding features to the custom games?  I really don't understand...

Good idea and makes perfect sense.  A person who is able to win most of his games as a pugger is definitely skilled, while a person who wins most of his games in premades more than likely has inflated stats.

They could possibly give you two different stats; ranked pantheon stats and custom team stats based on who you play with.  It would allow premades to have their own ladder that could more competitively rank them against other premades, and allow individual players and puggers to have their own ranks.

Or we can simply maybe not share everyone stats.

For example League of Legend only tells about the win and leave (leaver hehe) when seeing other players stats.

Of course, we can see the full stats of OUR character.

So what could be great is disable the online web page stats and only show the partial stats im telling you above in the connection info that is comming this week ?

Stats, as in Number of games, and win% are usefull, so os the familiarity %.

I will play with anyone if the teams seem to be balanced.  BUT it seems to me, premades especial, want to stack their team with people who have 80% win rates and play agianst people with 10% win rates. 

Since the overlay I have avoided countless pug stomps.  Its to the point of silliness.  There are so many premades who only want to pug stomp, and that give all premades a bad name.  Additionaly there are many random pugs who try to stack one side so it is a stomp.

To their credit (cant think of their names now) there are some who want an honest game, and I have had some good matches agianst those premade teams.