ReneDescartes wrote on 06/15/17 at 22:35:01:

Indeed. It seems like someone in FIDE has had a good think about how to prevent the excessive rating gains we've seen sometimes.

Ah I see-- because the system doesn't iterate the calculations after every game. That must then be the meaning of the statement that the  rating "turns over" in 18 games for K=40: If you play at a performance level 150 points over your last rating for 18 games you will gain those 150 points. Then the K=700/n business exactly prevents anyone affected from going over their performance rating, for it cuts the K so as to make the the totality of games equivalent to 17.5 games, just short of the number that would let them go over their strength.

Stigma wrote on 06/14/17 at 23:35:32:

Good point. In the US most tournaments are open and this is not such an issue.

If the probability curves worked as advertized, however, it would be impossible to overshoot your actual playing level except by fraud or collusion (throwing games). Playing at 2600 level for a year wouldn'the get you all the way to 2600. That doesn't mean that rising to your real playing level more quickly than can other improving players isn't gaining an advantage over them, but this advantage could not exist over established players--if, that is, the probabilities worked as advertized.

On the other hand, it's certainly imaginable that one can develop the ability to beat 2300 players with 3/4 of the points yet not the ability to overcome the qualitatively different resistance against scoring 1/2 of the points against 2500 players. Kasparov speaks of being stunned by the nature of the resistance he encountered in Petrosian and Spassky. His first Karpov match shows the same phenomenon. His rating after beating Korchnoi and Smyslov was higher than Karpov's, yet it did not prepare him for what he was about to meet. The swift result was 0-4... 

If this phenomenon is common then the proability curves underlying the rating system do not work as advertized but are deformed; in that case, one really can overshoot one's playing strength (or rather there is no single quantity representing one's strength in the current manner agains all players.

One could patch the problem with a rating ceiling of 100+ the rating of the highest opponent one has beaten, or average of highest 3 opponents, etc.  The current regulation would also help (but as I read the regulation, for juniors under 2300 all games in a rating period will have the same K: 40 or 700/[n for the period], whichever is less.)

There seems to be some changes...please see

"K = 40 for all players until their 18th birthday, as long as their rating remains under 2300.
If the number of games (n) for a player on any list for a rating period multiplied by K (as defined above) exceeds 700, then K shall be the largest whole number such that K x n does not exceed 700."

So what i understand is that k value is  700/ number of games played by the player..

am i right?  this is confusing because if a player plays say 100 games..then his k value becomes 700/100= 7  it doesnt make any sense...am i missing something?

ReneDescartes wrote on 06/14/17 at 23:17:19:

People who compete with that kid for tournament selections and spots on a team care.

This especially affects other juniors who have been less lucky and/or strategic, but also adults when regular (non-junior) tournaments are concerned.

It's high time this K=40 nonsense is ditched for players who are already way above 2000. The intent was good: to help juniors climb from very low start ratings after FIDE lowered the absolute floor several times. But the system in place for the past several years was an over-reaction.

I think its all been published, no big changes really.

https://www.fide.com/fide/handbook.html?id=197&view=article