Rating manipulation

Beetlby, the question is "why are high rated players forced to play such inequitable matches?"  The same thing happens at the 1300 level.  For example, in bughouse, I will get paired with a player rated <1000 (oftentimes 300), who doesn't know the game, doesn't communicate, and will resign even though I may have a winning position.  Because my opponent is also 300, I lose 25 points on the game.  If I'd won, yes, +1 pt.  So then, what's worse is when I can't shake that bad partner, and am continually paired - continually losing - and continually dropping points that I will struggle to recover.

... you can set a rating range, so you won't get paired with low rated players, or just very rarely. "

Your "solution" does not apply to the bughouse Arena or tournament events; the focus of my comments.

Well finding partners is another problem, as You said crazyhouse games power helped u, not individual rating of bughouse   To find partner has also simple solution: tournaments with changing partners, systems was many years ago already designed. And for such tourneys special rating should be also prolly, and such events can serve to find friends also.

Here is simple point and solution:

Main point: bughouse DOUBLEs Chess is game of TEAM, so in principle only team rating has sense. In real OTB tournaments You play often for victory so you will not partner some newbie, but some partner with which You will be stronger than simulin on both boards.  The simplest solution: team rating. So why anybody cares about single person rating ? If team play team only power of opponents as team is interresting info.

There is also a very good decent solution halfway to a team rating: The number of rating points played is divided among the players according to their single ratings, so that the greater part of the change is applied to the weaker one of teammates. Thus, for example, when playing for + -8 rating points, if the difference between teammates is 400, the weaker team's rating will change by + -14 and the stronger teammate will only change by + -2. While opponents after the same game if they have a rating in the distance of, for example, 100 points, they will both change by + -8. Real example: winners of 2020 bughouse championships made as team both about 3000 rating, if this method of calculating the change in rating after the games were applied, never could happen that few weeks later one of them fall to about 2500 by partnering much weaker partners, which wasnt able use his skills in teams battles.

I remember when started playing bughouse how interresting was games played for example vs legendary gnejs simuling, why remove good teams from server, let them play. Hunting individual rating in general has no sense in team game. But its important to explain it or better : to implement it.

Final question: is ANY chance this will be implemented here on chess.com or I just lost time writing it ?

Bughorse is a typical example that there is a great misunderstanding of individual bughouse double(s) chess ratings. It's funny what a contrast it is between the word single and double in one sentence.

Lets try also with proof by contradiction. 

If an individual rating measured the strength of an individual (similar to classic single board chess), it would mean that if there is a difference of 460 points between players, then they will beat each other in 95% of games. There are now a lot of players on chess.com (and it was always on all servers in past where implemented individual bughouse into double bughouse game as average of individuals) who fluctuate in time with a rating ranging from 300 to 500 ELO points. And among them are players from the weakest to the winner of the chess.com bughouse world championship. This means that if these players cloned themselves, they should play against own clone in a 9: 1 to 1: 9 ratio on 10 games. In other words, the granularity alias accuracy of the individual rating of the team game bughouse on chess.com is close to zero.

And now let's compare it with the expectations of ordinary people who know the rating from classic individual chess, where the accuracy is approaching 100 or 50 ELO points (ie the variance from 4.5: 5.5 to 5.5: 4.5 in 10 games), not zero, but on the contrary.

Any real argument against this?

Some people who dont know how funny the rating of individual in team game is can feel effect like DCMD2016 wrote :"My enjoyment has tanked so drastically" to whom nobody explained this. Chess.com doesnt come with warning: "dont take this rating so seriously as singleboard games ratings". The worst is a poorly explained virtual reality. We'd better focus on making game as fun and interresting as possible. Instead of spending time by focusing on something like individual rating in team game. And if there's a team that plays 2300 (as a team) don't worry too much about whether one has 100 and the other has 4500, or both 2300. Importantly, there's more about an interesting opponent for the game.  And if someone plays a simul on both chessboards, and has a 2300 team rating, there will definitely be a big challenge and fun to play with. Why ban it, just set the rating to the middle of the interval, ie 2300. Such an improvement could be made automatically for any team that plays exclusively together and it won't bother anyone, will it? 

"So you are saying that 1600 +1500 and 2300 +800 should be the same because both total 3100?"

How You know that somebody is 800 and somebody 2300 and not both arround 1550 ? 

Better is supposing both 1550 than 100 + 3000, or not ?

This one I didn't even understand, Is there a simpler way to explain it ?

Sure always should be way how to say it better to be more easy to explain. Although You can never go with elephant thru very small doors.

What about this:

Well ratings should be based on skills. Its question which skills. Lets call skills ability to win, so in present online including also having low lag and good partners. Also this ability include to advice moves to partner (on higher level its about coordination so You have to communicate lines possible fast to make parallel  processing effective enough) i.e. playing simultaneous on both boards. So if You can for example play on both boards like 1550, then definitely You are not in team worse than 1550, probably You are in team (if You cloned = You will find partner similar power and style)  at least 1750 or more. 

Or vice versa: here is bottom limit of team power : its the simuling (playing on both boards) power of stronger player in team. If we supposing that team communicate. Possibility of communication in team  leads to effect that if one of players in team is really much weaker, then optimal distribution of parallel processing should be like this: all the time stronger player playing all moves on both boards (resp. saying what to play) and only for example in last seconds or so, the much much weaker partner (player in team) will play something to prevent team losing on time. So if 1900 simuling will partner 100 simuling , definitely will team power be not 1000 ( (1900+100)/2 ), but around  1900.  And this "inflation" between real team power and nominal team power calculated by (player1+player2)/2 should not be added to better player, simply because its evaluating his ability to find somebody who is as weak as possible in bughouse chess, but is not as weak in making moves which stronger partner will tell him (So he has functional ears). Its why the "profit" from simuling should not go to stronger player in team , but mostly to weaker, because we like to helping weaker players by advice them moves make them stronger (develop their skills in long run), but we don't wish for it pay stronger players by rating points drastically. Ratings are not money, they should measure the strength of player.  People feel that way , its why they don't like simuling player has 3900+100 , but are fine with 1900+1900. The solution offered above (team rating, or at least modifying more rating of much much weaker player in team) should be applied to make things right.