Is chess.com average rating is 1200?

Our chess.com must have started with 2 players, both with rating 1200. One won, other lost. Average must have remained 1200.

Extending same logic, so even now, chess.com's average rating should be 1200 - right?

Not quite....

some people join the site lose some games and never return or close their accounts. In this way rating points migrate from dead accounts to live accounts and the average rating rises above 1200:

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

Players:	75751	Rating Distribution
Average Rating:	1384
Average Glicko RD:	133
White Wins:	N/A
Black Wins:	N/A
Draws:	N/A

As you can see the average rating for Standard (turn-based) is in fact 1384.

Suppose a new member "A" plays his first game against a player of rating 1500 with 10 games under his belt and beats him.

Now suppose another new member "B" plays his first game against a player of same rating (1500), but with 100 games under his belt and beats him.

Will A's increment in rating be same as B's ?

Very good thinking!

Okay, here's my thinking (based on logical thought {I hope} and with no reference to the formula that governs this particular calculation):

Assuming Players C and D have been playing regularly and recently (also assuming various other factors are equal i.e. strength of opposition):

Player A beating Player C (10 games: therefore higher deviation from true playing strength) gets less points than Player B gets for beating Player D(100 games:  therefore lower deviation from true playing strength).

It's the only decision that would make any sense to me.

For a definitive answer:

http://www.chess.com/article/view/chess-ratings---how-they-work

just play chess...

 I agree;)

I think glicko's RD provides more to avarage rating not being at 1200 than anything else, but I could be wrong.

http://math.bu.edu/people/mg/glicko/glicko.doc/glicko.html