Max. boot rate, min. games played, blacklist: these are all indirect ways of getting good players in your game. One of two things could be done:

1. Have a min. winning % for the type of game proposed. Eg: I create a 3 v 3. I make the min. winning % to join 55%. Only players who meet this requirement can join.

2. Players are given a rating (like pogo.com games or warlight's ladder). I beat a good opponent/team, my rating goes up more than if i beat a bad opponent/team. Vice versa for losses. Then, a minimum rating could be used to limit people who join. Eg: good 1 v 1 players on the ladder seem to be above 1500 or 1600. Use those numbers as minimums to join.

Vote at:

http://warlight.uservoice.com/forums/77051-warlight-features/suggestions/2229310-min-winning-to-join-and-or-rating-system

well theres already a minimum rating thing for the ladder values. and obviously thats a member-only feature.

For games like cribbage on pogo.com, it is a basic feature for all. The ladder is competitive in nature. A simple rating without rank would serve a different purpose.

Many good players I know do not care about their win%. They join random games to play for fun. In doing so they are able to teach teammates how to play better and everyone in the game has a chance to review history. Ironically that is exactly how I learned to play. With your method it would be expected to see good or dece players joining games with bad players just to inflate their win% to join your games (or others with higher min win% prerequisites) ATM there are much better ways to spend your ten votes (Global chat/whos online, Profanity filter, Soft boot)

you must care about this game to much if you only play to win it

the reason why a simple rating system does -not- work for warlight, and it does for cribbage on pogo.com... cribbage games are -always the same- every game follows the same rules, and each player can be rated on their gameplay because of that.. in warlight all features are entirely customizable.. if you wanted to inflate your w/r then you could just make some impossible features and mislabel it incorrectly.. or sadly use an alternate account to falsely increase your rating.. any sort of a non-ladder sorted ranking system, would be almost as useless as the winning ratio.. Imho of course..

Winning percentage is essentially overrated. My own 1v1 % is getting lower now because I mostly play my games in high-level tourneys after having piled up wins in easier auto games with random opponents. Other players have older IDs or IDs created after their learning period and only played against strong players, and these can be crazy strong themselves although their win rate is not fancy at all.
Any really good player will have, for say, above 50% wins, and except sandbaggers or, well, any kind of cheater (see what Perrin said), anyone above 80% has to have understood more of just the basics of the game. Still, it doesn't tell you much more about skill.

Actually, I say this for one simple reason: I prefer 3 v 3 games with good teammates and good opponents. It has nothing to do with inflating anything. I'd like to be able to create 3 v 3games with min. winning % of 50-55. That's all. Many players care so much about winning % on their stats that they don't join team b when team a is good. With a min. winning %, they would be much less hesitant to join. Games would be more competitive.

i have same motivation as Yuanshuai - playing in strong (realtime) games is great.
often i have less fun because my teammates are both weak players AND unwilling to take advice, or even work together. i have found that almost all players with good win % will cooperate with teammates, and that makes games fun.
best would be a (crude/imperfect) rating system, as win% is really not such a reliable indicator. also a rating system would encourage players who worry about their win% to still play on teams with weaker players (because losing would not damage their rating much). the TrueSkill algorithm would be a good one to use, maybe?

|> Many good players I know do not care about their win%. They join random games to play for fun. In doing so they are able to teach teammates how to play better and everyone in the game has a chance to review history. Ironically that is exactly how I learned to play. With your method it would be expected to see good or dece players joining games with bad players just to inflate their win% to join your games (or others with higher min win% prerequisites) ATM there are much better ways to spend your ten votes (Global chat/whos online, Profanity filter, Soft boot).

I have to disagree here BRILLIANT, i find that very few of the better players join random teamed games. Personally, i think you should have completed single player levels 1, 2 & 3 to be eligable for ranked games. Some players i end up with have no basic knowledge of the game.

I dont know whether i am a good player but i really dont pay attention to my win%, especially in team games. I join them and just want to have some fun, no matter whether i loose or not

I think for every player like brilliant's friends, retro or jacas (a good player I know who joins any game for fun), there are 5-10 others who pick and choose the teams they join mostly based on perceived overall team strength.

If I saw a game whose name was "Europe 3 v 3, 55% win min." I'd join without question bc I'd know I wouldn't end up with someone on my blacklist as a teammate.

Another option would be to have a game setting where it balanced teams in the beginning of the game based on perceived strength (at this point...win %, although it's a bad indicator of skill for reasons people have stated already). So if 6 players joined a 3v3, it would attempt to balance the skill of all the players when determining teams.

I don't play many real time games these days, but when I used to, I would not join a team with poor players or new players if I saw that the opposing team was stacked. Why? Because lopsided games aren't fun, and the point of playing is to have fun. Balancing teams out would avoid a lot of the lopsided nature of games.