What do you do when your rating is too high?

I don't expect an 'accurate' rating, just one that is fair! Now it is even worse.

Well if you didn't beat everyone you played, you wouldn't have these problems

 True! My opponents keep picking Openings that I have been studying and know fairly well. I also have been studying how other players play these openings. So I am getting to the Middle Game without making any major mistakes. After that, if my opponent has made a mistake, I am up both material and position: so as long as I don't blunder (and I have!) I am doing OK.

In the initial stages of establishing a new rating, there have to be large and wild swings back and forth in the rating for the first few games until enough of a database is established to begin to have some reliability.  It's not a matter of fair or unfair-- that's just the way the math works.  Some systems label everything as a provisional rating until a minimum number of games, say 20, have been played.  Maybe that's how you should be thinking of your early rating-- ignore it and pretend chess.com won't even give you a rating until you've got several games contributing to it.  As an indication of ability an early rating, especially after only 3 games, is meaningless.

So far you're still winning, yet already worried that now you'll be forced to play against opponents of too high caliber. Your idea of "competitive" seems to be people you can beat on a regular basis.  If you play against truly competitive opponents, you'll lose half of your games in the long run.  If you really don't like losing, as you said, you either should not play chess or play only against little kids, and even then one of the little buggers will surprise you once in a while. 

Your RD is still pretty high (192).  You've completed just 3 games so far.  I wouldn't worry too much about the rating just yet.  :) 

My idea of competitive is where I win approximately half of the games I play and lose about that. I don't like losing but losing is a necessary part of developing better play. It seems that I remember my losses more than my wins.

I am not certain what RD means. I will have to look that term up.

EDIT: OK, now it makes sense to me:

"These numbers are found from the cumulative distribution function of the
normal distribution with mean = current rating, and standard deviation = RD."

I wasn't certain how ratings were set or or changed. This is perfectly logical. Thank you for directing me to it. It certainly indicates that my rating has not settled to a narrow confidence region.

60% or more of your games being wins is competitive! 50%? Normal? I don't know

I think maybe it makes a difference which direction one looks from.  I was referring to whether the opponents are considered to be competitive-- I'd say, long term, if you want to claim your opposition is competitive, they should have an (average) abilty equal to yours and equal chances of winning.  If someone wins 60% of their games, then they are playing people who, on average, are weaker than they are-- not competitive.

But from the other direction, as the term is used in general speech referring to the success ratio of the protagonist, I agree the way you state it is a common approach.  I think in sports the term "competitive" is often used almost synonymously with "successful".  If a player says "we're going to be competitive this year" s/he doesn't usually mean "we expect to win half our games and miss the playoffs!" 

My apologies, Cystem, I was referring to the original poster's wish to win half his/hers games. But, as always nice point

Good points!

I would say that you want to win all of your games that you have with weaker opponents and some percentage of wins against stronger oppponents. Of course you will have losses against some weaker opponents but the majority of those games should be wins. I don't know how to quantify it better.

I did not think of ratings as a mean score of a distribution but I should have. And as a seasoned researcher, I should have also realized that a mean is worthless without a measure of distribution (the variance, standard deviation, etc.,) and the shape of the distribution.

For turn based chess, the shape of the distribution is at

http://www.chess.com/echess/players.html

A normal distribution is not a bad fit although there is a long tail for higher ratings.

 Yup, now you got it!

I was looking at your game against Stoned Eagle - you played a lot stronger than your average 1250 rated chess.com player. I think you underestimate yourself.

 Thank you for your comment.

I did make a few mistakes but no obvious blunders (although during the game I thought was.) I did not always play the strongest moves. In my post mortem, I picked up a couple areas were I could have done better. But thank you.

I thought that you created the opportunity to pin the queen and it did snare your opponent. Then isolated the pin as a weakness and put constant pressure on it. So in my eyes you used a good tactic and followed through with a clear plan. Honestly puts you higher than a 1250 player :)

What improvements did you see?

When my rating gets high, I take on the big boys. They always put me back in my place.

Maradona,  after every game that I play I do a post mortem. Here is my post mortem from the game (please excuse my bad spelling, when i did it, it was only for my benefit):

Do you agree?

q:What do you do when your rating is too high?

a: become a GM lol

seriously can a rating get to high. your rating will automatically adjust to your playing strength over a period of time.

Hahaha.  I'm not sure if you were going for laughs with that, but it is funny.  Thanks for the laugh.

Your rating won't change but to get over that hurdle try playing unrated turn based.

Poor Soach never got that loss he was looking for to make his rating lower and more "accurate". He hasn't logged on in more than 6 years, but he finished up with a grand total of 12 daily ("turn-based" as they were called back when he started this thread) games, a record of 11W 0L 1D, and a rating of 1874.