huge rating jump

I see. So the numbers from the diagram still reflect the previous rating distribution, where only a handful players are >= 2100.

Yes, as you can see the standard ratings clearly had a problem.

Agreed with Bigpoison. For instance, there's no one between 2100 and 2200 - if I hadn't played my last two standard games, I would now be 2205 while my actual std rating is 2088. The algorithm should have been linear, something like [new rating = 1.15 * former rating] instead of adding points in a very rough way.

Apart from this, std ratings tend to be now rather inflated, while they were deflated before. All of this was basically mindless.

Yeah that's what I said, and I agree it would be far more accurate.

But this is easier which seems to be the priority. Ratings for some people will wildly inaccurate unless they play a number of games quickly.

Another problem is that we now have a gap in the 2000-2200 range and anyone 2200+ has no incentive to play anyone below 2200 because they know that their comparative strength to someone 2200+ will be really high.

It's really a horrible solution to a minor problem.

And some have their rating unchanged? Is that also a glitch?

I play on several chess sites and have always found the players here to be stronger than elsewhere, especially lower rated players, so I think all in all that the adjustments are for the best.

I'd love to know what the 'various factors' that caused the deflation actually were.

Mine is unchanged as well, and I'm guessing it's a glitch, otherwise I've been sandbagged :)

Tell Fide that you are 2300++ on chess.com maybe they'll give you one

monitor, would you mind going into the nuts and bolts of what changed in the algorithm? or is that proprietary info?

I geek out on this kind of stuff. 

I did notice that some real life Masters were rated relatively low in standard chess which got me wondering. But I do not think that most people below 1500 or so were underrated, so I dont know if that was the right approach to give everyone a higher rating.

How did this rating deflation happen? was it because people cheat with chess software which skews the ratings and when they get kicked out the rating points taken from their enemy is lost forever too. This is probably more of a problem in standard timed chess since the less time is given the less likely is the use of a chess engine.

The algorithm remains the same, and after a few weeks of everybody playing each other the ratings discrepancies created by the difference of adding 300 or 400 will dissipate so it is unimportant that currently some players are underrated or overrated.

If they play against each other they will resume their level and if they sit on their level then they don't form part of the pool anyway and we can forget about them.

The end result should be a shift in ratings on the high-end where we will start seeing regular legitimate 2000 vs 2000 level games, whereas before that was incredibly rare.

Strange how I jumped literally  500 points, from that 2000 range to 2500...

not sure how that worked, Ithought I was only supposed to geet like 300?

So have you fixed the underlying causes of the rating deflation, or are you just going to change the ratings again unannounced in a few years?

Congratulations GM chesskingdreamer 

              Too bad chess.com can't do that with my bank account

The cause of the rating deflation started right at the initiation of the pool; we've never had realistic ratings in standard.

The pool is already in place now with tens of thousands of regulars, so the problem should not occur again.

Why couldn't you communicate such a major change that you knew would cause confusion?

They can, if you don't mind the sign also changing.

cool, now my standard rating is 2300. Can I get FM title now?