How will this effect the graphs for us premium members? Right now the ratings still haven't been updated.
huge rating jump
Sort:
I'd like to say thanks to chess.com for giving me the highest rating of my career!
hahahahahahahahahaha All right, you've had that rating for long enough, NOW GIVE IT BACK
So... if the formula couldn't differentiate players of different skill for... some reason... then isn't this just a temporary fix?
Oct 21, 2014
0
#126
What a lot of people aren't getting is that there was a logjam of players in the 1850-2150 range. What this will do is open it up so that there will be a real difference between players. Instead of all the top players crammed into a 300 point range, they are now spread over a range from 2000-2750.
This will actually make the ratings more accurate in the long run.
SmyslovFan wrote:
What a lot of people aren't getting is that there was a logjam of players in the 1850-2150 range. What this will do is open it up so that there will be a real difference between players. Instead of all the top players crammed into a 300 point range, they are now spread over a range from 2000-2750.
This will actually make the ratings more accurate in the long run.
If players are really that far apart in ability, they wouldn't be so close in rating now. If someone was 2,200 and not increasing it's because they don't score any better than they should against the 2,100s and so on.
Mind if I spread post 133 around ??
Oct 21, 2014
0
#129
sorry to say that, but this whole action with the rating changes was COMPLETE TRASH!
Till_98 wrote:
sorry to say that, but this whole action with the rating changes was COMPLETE TRASH!
1616? Noob!
Oct 21, 2014
0
#131
Scottrf wrote:
SmyslovFan wrote:
What a lot of people aren't getting is that there was a logjam of players in the 1850-2150 range. What this will do is open it up so that there will be a real difference between players. Instead of all the top players crammed into a 300 point range, they are now spread over a range from 2000-2750.
This will actually make the ratings more accurate in the long run.
If players are really that far apart in ability, they wouldn't be so close in rating now. If someone was 2,200 and not increasing it's because they don't score any better than they should against the 2,100s and so on.
Well, I'm currently 2068, and some of the people who were within 100-200 rating points yesterday are now 2300 and 2500.
Till _98 your rating changed to 2116, but is still wrong. According with the rule, your rating must be 2216. ratings > 1800 are boosted with 400 points.
smyslov, you rating must be 2268. Seems that the program makes some mistakes.
Oct 21, 2014
0
#134
By giving titled players an extra 100 rating points, it spreads out the rating distribution.
SmyslovFan wrote:
Scottrf wrote:
SmyslovFan wrote:
What a lot of people aren't getting is that there was a logjam of players in the 1850-2150 range. What this will do is open it up so that there will be a real difference between players. Instead of all the top players crammed into a 300 point range, they are now spread over a range from 2000-2750.
This will actually make the ratings more accurate in the long run.
If players are really that far apart in ability, they wouldn't be so close in rating now. If someone was 2,200 and not increasing it's because they don't score any better than they should against the 2,100s and so on.
Well, I'm currently 2068, and some of the people who were within 100-200 rating points yesterday are now 2300 and 2500.
Non sequitur. They've boosted different players different amounts (sometimes by accident, sometimes on purpose - titled players) and it will depend on how many games and who they have played against since.
Oct 21, 2014
0
#136
jaadlh_6691 wrote:
smyslov, you rating must be 2268. Seems that the program makes some mistakes.
You're probably right, but 2068 is similar to my USCF ratings when I was active. I'm not complaining about my rating.
dragoscscc كتب:
dragoscscc كتب: Same here, 1388 to 1538
Same here, 1388 to 1538
@Smyslovfan
At this post, you have the increasiing rating rule
http://www.chess.com/forum/view/livechess/standard-ratings-boost?page=1
Generally, the rating (at chess.com) of the titled player is different. If you see the games, at one point changes hardly. I think they get the official rating (OTB) they have or something like this. Probably, it is not logical that a titulated player starts with 1200 rating point as patzers like me, for instance. But sometimes they have this special rating and another times, the usual one. I read at another forum that people from the deceased yahoo chess, they manteined their yahoo ratings, but I don´t know if that thing is true or not.
For me, it is not a real problem (speaking at medium term). If whatever player (titled or not) have a not real rating (overrated or underrated, no matter that), when this player starts to play, this player is going to arrive to their real rating (according with this strength). I think that the glicko system is technically superior than Elo (because of RD factor) and like the people here use to play a lot of games (more than OTB games, because it is easier to play online), the rating is reflecting better the real strength of the player that OTB ratings. Obviously I am not including at the equation the cheaters, but I think that the "rating suprises" at a game are hardly more difficult here that an OTB tournament. Simply because a rating made with 300 games is more representative than a rating made with 50 games, for putting and example.
First mistake in this reshuffle: those of us who haven't yet received our boost now have very happy hunting among the boosted who are now our similarly rated opponents. So we can go on a binge, win a lot of games, and when our boosts finally come we shall have a much greater net rating rise than if we'd had the boost along with the first wave. This will again warp the ratings, so we'll all probably get another boost or rating cut to even things out again. Looks like providing forum fodder for some time to come.
This thread is about 12 months old, but it may help some new members http://www.chess.com/forum/view/general/fide-ratings-vs-chesscom-ratings-explored?page=1