rating system

did the rating system change cuz now im not going down as much as before when i lose and im going up more when i win, also my opponent doesn't go down as much as i go up when i win and vice versa... its hard for me to tell if im actually getting better now cuz i go up as long as i win about half of them

the rating system is fugging bullshit (sorry).
i´ve played over 14000 games so far, my rating was always between 1700 and 1800. few day`s ago, i won almost every game, so my rating is now over 2000!!! some other (human) player feel the same thing. there rating rising up from 1700 to over 2000. But we dont play better than before! I suppose on chess.com there are many virtual players (computers). is there a bug on the virtual machines?

the next problem is that players which have a high rating (for example 2000 and more) plays only against opponents which have a similar rating. so the rating system have no explanatory power. the rating system would have more informative value when all players are forced to play against "light" opponents.

this relates closely to my experience. I used to float around late 1400s. stop playing for a while and watched chess videos and put about 150 hours in tactics training -  and now that I'm back my rating is almost 1600. to my surprise and delight I won a a few games today against late 1600 early 1700 players. while the Hard computer at 1750-1800 still blunders once a game or once in two games and I almost alway beat it. (when I used to play there was no impossible computer...

I'm just wandering how would this rating compare to real life rating like FIDE rating and whether it is reliable and not inflated/distorted. Oh and to mess things up even more I've come across player who are masters an have a rating of late 2000s early 2100s - which is lower than the usual 2200 for masters and also the complete opposite phenomenon which I fear may be happening at my level of play: rating inflation

Any inputs will be appreciated

same here, was 1100, did no studying on chess and rose my rating 400 -500 point higher

The ratings were deflated so we made a small adjustement to them.

 Really? My chess.com rating is about 1250, yet at a local chess club where I play where they rate according to FIDE, my rating is around 900.

The rating of any pool of players is relevant to that pool only. A chess.com rating has nothing to do with FIDE.

The site can decide to decrease or increase the numbers, on a whim or for valid reason, for everyone and it would all stay relatively the same.

For instance, if the starting point for new players was 12000 rather than 1200 your revised rating of 12050 would be as relevant to the chess.com pool as it is now.

Meanwhile, everybody's rating is wrong as the most active players ratings will begin to inflate while the less active will stay deflated. Great, now the ratings aren't even comparable within the chess.com pool due to this artifical flux. Then the less active players ratings will catch up over some period of time creating a second wave of artificial flux in the ratings. And at no point do you know if the next player you play will have an inflated or a deflated rating. Not to mention using your rating as a benchmark for progress is out the window. Oh well.

In what sense were they deflated??

 Leave them deflated! Then you play in a tourney, beat 2000 rated players, think it'sa fluke, and do it again! Then you get all happy, and it's so much fun!

Well it's just an idea...

yeah, now instead we r going to go play in a tourney and beat 2000 players and think "well thats expected"  or lose to them consistently and say wow i suck

I don't have much to say about the rating system execpt- how in hell do i beat someone 400 points higher than me and only get 16/17 points while he loses about 40!?! Why wouldn't I gain more than that? I don't see how that is fair at all! Then again the two times he beat me I lost just one point... But still I don't understand how this works.

I can see how ratings can deflate.  A strong player joins chess.com and slurps up rating points from his early opponents until he reaches a stable level.  Or a weak but rapidly improving player does the same thing.

The USCF handled it with bonus and feedback points, but they had a whole tournament to rate at one time and it was easy to see when a low rated player did way better than his rating would indicate.  Rating games here one at a time makes this a lot harder.

Sounds like he had a higher K-factor than you. K-factor is the volatility level of your rating, and it goes down as you play more games - so your rating changes a lot when you start off, and gradually settles down.

At least, this is how FIDE ELO works, and I'm assuming Chess.com is the same.