this game is dead

I would argue that even people deciding to attack the higher rated player is part of it and is included in your edge.

But I can see how that might make it so that your goal of having fun is compromised in a given individual game when your rating is very high.

I totally agree and I said a few times on this forum that the matchmaking now is far from ideal (to say the least). And other players also said that. I believe the developers will fix it.

Now imagine that they fixed it and 1900+ players will play with others from 1800+ to 2000+ only. Will you still say "this game is dead"?

@babyagun 

I guess not.

I know, play and enjoy most of the expansions (variants), but they are not meant to replace the standard game, which remains the most popular anyway. Also picking cards at random involves no less luck than throwing a dice. You know beforehand the distribution of the cards in the deck, as you know beforehand the distribution of the numbers on the dice. The point is that a lucky card pick or dice roll can decide the winner of a game, and often does.

Made this screenshot a minute ago:

1290 vs 1350 vs 1940 vs 1980. The difference between the weakest and the strongest is 690, and this is usual. We definitely need a better matchmaking algo.

Another thing about wraku is that he is a GM with a FIDE rating of 2500, so those who know that won't fancy their chances against him once it gets down to 2 players. Thus they will try to take him down early. 

Can't improve matchmaking unless players want to wait even longer for matches and we've already gotten tons of complaints about how long it takes.

@battleMind24, please make an option. I am ready to wait longer to get better match. So, I will turn this option ON. If someone does not care about, he will have it OFF and play in a match like 700 vs 1200 vs 1700 vs 2100. It is up to them. But it should be the player's choice.

Regardless such an option would make it take longer for everyone all around.

When the rating difference between the highest rated and lowest rated is over 500 I consider it really bad. I would rather wait forever or abort such games instead of playing them, although I never abort based on ratings (when abort was still a feature) since it would be annoying for my opponents to get an aborted game.

There was a time where I noticed that the longer I waited the worse the pairings could be, so I have cancelled me entering the queue and reenter. Not sure if it actually worked well or if I was imagining things.

One example that might make pairings easier is to create a rating cap and rating floor when it comes to making the pairings. Like treating all players with rating over 1900 as if they had 1900 rating and treating all players with rating below 900 as if they had 900 rating. Something dynamic like the #50 ranked players rating from the top could be used for the rating cap or the #10 ranked player online. Not sure how well it would work, but just an idea to make it easier to find reasonable pairings at the top and bottom of the rating list.

Do you have a better solution? Let's discuss it.

Now many top players stop playing, because there is a risk get matched with 3 1200+ players and lose 100 points in a single game. It is not ok.

The current situation is discriminating. We see a "whining driven" process here. A crowd of whining people sends their complaints to chess.com team and the team does something just to stop the whining. And that "something" is not the best solution, it is just a quick workaround. Unfortunately the workaround then may stay for long long time. Please do not fall in this trap, @battleMind24.

I think the fundamental problem is using the math of the 2-player-chess rating system without adjusting for the "luck/randomness" aspects of the game.  This gets a bit into probability theory, but the basic idea is "the more surprising an outcome is, the more it should impact our assessments of probabilities".  In 2PC, a 1400-player beating a 2000-player is incredibly surprising; major adjustments should be made.  This is in keeping with intuition: a 1400-player that can consistently beat 2000+ players ought to see their 1400 rating sky-rocket to 2000 quickly; a 2000-rated player unable to defeat 1400 players ought to see their 2000 rating drop precipitously.

But in 4PC, a 1400-rating player finishing ahead of a 2000-rated player is hardly surprising at all.

I think a simple fix would be to use the same math as is used today, but divide final adjustments by something like a factor between 2 and 10. 

Going one step further, I think it's safe to say that the greater the gap between highest-rated and lowest-rated player, the greater the chances are for "shenanigans" ("let's get the high-rated guy!") that reflect very little on players' skill levels.  So it might make sense to use a rating-adjustment factor that takes this into account.

In short form:

Final_rating_adjustment = (original_rating_adjustment) / (2 + ((max_rating-min_rating)/100)))

@Martin0, you do not understand. The 4 Player Chess developers are very good programmers. They do know how to create a good matchmaking algorithm. But they do not want to change it. Because the new algo may potentially cause a longer waiting time. And this will increase the amount of complaints from impatient and eternally unhappy users.

Chess.com team seems to consider bad matchmaking as a lesser evil. Top players can not send 100's of complaints daily. So, their interests are not going to be taken into account. At least it looks this way now.

@kevinkirkpat, there is a -400 to +400 cap in classic chess on chess.com. And as I remember, it is -200 to +200, by default. If you are 1400 you normally play with 1200 to 1600 opponents. And you can not play with a 2000 player if he is not your friend. Because a 600 point difference is too much. Why don't we have any limit in 4 Player Chess?

Don't get me wrong @BabYagun a cap is a good idea right now, but when we add Teams the available users in each pool will be half in which case it would be senseless to do that as matchmaking times would increase dramatically (already ran a few tests). What's the best thing to do then? We need more users.

We can't compare a 40,000 - 60000 matchmaker to a 600 user matchmaker, as that makes no sense. Not to mention even in standard the highest rated only play against opponents they choose, something we cannot allow at the moment.

A topic related to this discussion: Possible solution to "let's get the high-rated guy!"

@battleMind24 maybe a cap that just prevents the most extreme imbalances would be a good compromise for now?

This game is a bomb and the users are going to come. People might whine that it takes too long but I don't think after the fact they are going to have a bad sense of the user experience for having had to wait 1 minute for a game that then lasts 10-25min.

Those situations with a really huge rating difference make for a bad experience. When I start a game and I have 3 players at -400 I am not happy. Can definitely understand people in the leaderboard not playing. I imagine it's also not nice for a low rated to be matched with much higher rated and get mated in 3.

For what it's worth I wouldn't prioritise adding teams. Great work by the way, thanks for the game 

Okay guys, you've convinced me I'll set a cap of 400, in other words, +-200 your rating.