any idea what a 1950 rated chess.com rapid player would be on fide

Sort:
Oldest
DCthedestroyer

If anyone has been at this level, please tell me

tygxc

1950 chess.com rapid is about 1900 FIDE rapid.

medelpad
Probably like 1700 FIDE
Fetoxo
1800
tygxc

More precisely: 1950 - 38 = 1912

# Player chess.com FIDE difference
1 Carlsen 2896 2823 73
2 Caruana 2757 2722 35
3 Nakamura 2769 2746 23
4 Abdusattorov 2770 2733 37
5 Giri 2741 2687 54
6 Ding 2719 2818 -99
7 Firouzja 2829 2724 105
8 Nepomniachtchi 2811 2754 57
9 So 2843 2742 101
10 Wei Yi 2740 2743 -3
Average 38Standard Deviation 59

tygxc

@10

Those 10 grandmasters have an average of 352 rapid games on chess.com with a standard deviation of 199. That is enough for a stable Glicko-2 rating.

# Player chess.com FIDE difference games
1 Carlsen 2889 2827 62 214
2 Caruana 2747 2738 9 499
3 Nakamura 2762 2746 16 779
4 Abdusattorov 2759 2724 35 473
5 Giri 2732 2679 53 339
6 Ding 2719 2775 -56 125
7 Firouzja 2867 2724 143 202
8 Nepomniachtchi 2813 2754 59 258
9 So 2827 2722 105 442
10 Wei Yi 2740 2771 -31 185

Average 40 352
Standard Deviation 59 199

Thus 1950 chess.com rapid = 1950 - 40 = 1910 FIDE rapid.

medelpad
#9 You can’t just use 10 top player accounts of which none of them play online rapid outside of infrequent events. Even in that small group the rating varies from +101 to -99.

For it to be applicable to someone like the OP there would need to be a much larger sample size including players who play frequently online and OTB.

It would also be useful to not only have an average but also the median and the type value.
CastPoc
tygxc wrote:

1950 chess.com rapid is about 1900 FIDE rapid.

I'm sorry, but there is no way that is accurate. I'm 1850 rapid, 1900 peak, and there is absolutely no way I'm even close to 1900 fide. My fide is 1480, but my actual fide rating is probably a tad bit higher because I don't play over the board that much.

Someone else said it's about 1812, but this is likely pretty inaccurate too. The only 1900 rated person I've seen with around 1900 ish fide rating was like an old guy who hadn't played otb in decades. I don't know, I may be wrong, but try and find a 1950 ish rated active player with a fide rating of 1800+. Probably not a lot out there.

1950 rapid is probably around 1650-1700 ish fide for someone who regularly plays fide rated events.

DCthedestroyer

First of all, my FIDE rating is 1516 since I have not played any events in a long time and will play after 2 months

Also, in a place where there is more competition, its harder to increase rating.

For example, 1516 was my rating in India due to better competition. However since I have moved to Ireland, where the competition is lesser, I am able to beat 1750 - 1800 fide people in my chess club.

Please correct me

Rapid_Chess_Only
CastPo wrote:
tygxc wrote:

1950 chess.com rapid is about 1900 FIDE rapid.

I'm sorry, but there is no way that is accurate. I'm 1850 rapid, 1900 peak, and there is absolutely no way I'm even close to 1900 fide. My fide is 1480, but my actual fide rating is probably a tad bit higher because I don't play over the board that much.

Someone else said it's about 1812, but this is likely pretty inaccurate too. The only 1900 rated person I've seen with around 1900 ish fide rating was like an old guy who hadn't played otb in decades. I don't know, I may be wrong, but try and find a 1950 ish rated active player with a fide rating of 1800+. Probably not a lot out there.

1950 rapid is probably around 1650-1700 ish fide for someone who regularly plays fide rated events.

Your fide rapid rating is 1480 or is that your classical rating?

tygxc

@16

"I'm 1850 rapid" ++ Chess.com, then 1850 - 40 = 1810 FIDE rapid

tygxc

"They don't play enough rapid games for their rapid ratings to be... ACCURATE!"
++ 352 rapid games is more than enough for an accurate Glicko-2 rating.

Rapid_Chess_Only
andblunderstheking wrote:

@tyxgc you need to agree you are wrong , in my country a 1900-2100 can be anywhere 1400-1600 fide.

From an outside perspective, it appears that most people who have responded to tyxgc are conflating rapid and classical ratings.

Silent_Assassin

Well, for a more accurate rating we should check the starting ratings for a beginning player. Before it was 100 FIDE, but today, we can say that—

1200 chess.com = 1500 lichess (according to starting ratings

1200 chess.com/1500 lichess = 1000 FIDE (Previous standard), 1400 FIDE (Today's standard)

So, how do we actually identify? Well, both are the same due to the 400+ adjustment introduced by FIDE in March.

But, as the lichess ratings go higher eg. 2100, Chess.com gets lower.

+2000 = every extra 50 lichess points, difference increases by 25.

eg. 2100 lichess = 1850 chess com.

Through this difference, we can also say —

1950 chess.com = 2150 Lichess = 1900 FIDE.

"1900 FIDE is your estimated score as measured with the same amount of your chess.com games."

But remember : chess.com is USCF and FIDE, Lichess is French. You need to do one more calculation for a greater accuracy.

USCF = 800 = 1400 FIDE. Average = 650 among them,

= Your final rating is 1780 FIDE.

Silent_Assassin

I believe this is the correct answer so yes. My theory and 🧠 🧐

Silent_Assassin
DCthedestroyer wrote:

First of all, my FIDE rating is 1516 since I have not played any events in a long time and will play after 2 months

Also, in a place where there is more competition, its harder to increase rating.

For example, 1516 was my rating in India due to better competition. However since I have moved to Ireland, where the competition is lesser, I am able to beat 1750 - 1800 fide people in my chess club.

Please correct me

This is due to the increase in rating inflation. Players playing online chess increases and their levels as well. They appear and perform suddenly extremely well and you might think that it's their rating and you are under the level. But yes, isn't that why the 400+ thing came into play?

Plarrer

According to chessgoals, your rating should be around 1925 fide

Rapid_Chess_Only
Silent_Assassin wrote:

Well, for a more accurate rating we should check the starting ratings for a beginning player. Before it was 100 FIDE, but today, we can say that—

1200 chess.com = 1500 lichess (according to starting ratings

1200 chess.com/1500 lichess = 1000 FIDE (Previous standard), 1400 FIDE (Today's standard)

So, how do we actually identify? Well, both are the same due to the 400+ adjustment introduced by FIDE in March.

But, as the lichess ratings go higher eg. 2100, Chess.com gets lower.

+2000 = every extra 50 lichess points, difference increases by 25.

eg. 2100 lichess = 1850 chess com.

Through this difference, we can also say —

1950 chess.com = 2150 Lichess = 1900 FIDE.

"1900 FIDE is your estimated score as measured with the same amount of your chess.com games."

The reason for the relationship between lichess and chesscom ratings is because of the rate of change for wins and losses. On this site, you are awarded 8 points for a win against an opponent of the same rating, on lichess you are only awarded 6 points. If you put it on a graph, you'd see that you could predict exactly at what level a player's expected rating on chesscom would surpass their expected lichess rating. Fide awards ~5 points for a win against a similarly rated opponent so I would expect that you would need a 2000 rapid rating on chesscom to be similar to a 1900 fide rating.

The only problem is that this is only an estimate because there can be no exact conversion between two different pools of players.

Rapid_Chess_Only
Optimissed wrote:
Rapid_Chess_Only wrote:
Silent_Assassin wrote:

Well, for a more accurate rating we should check the starting ratings for a beginning player. Before it was 100 FIDE, but today, we can say that—

1200 chess.com = 1500 lichess (according to starting ratings

1200 chess.com/1500 lichess = 1000 FIDE (Previous standard), 1400 FIDE (Today's standard)

So, how do we actually identify? Well, both are the same due to the 400+ adjustment introduced by FIDE in March.

But, as the lichess ratings go higher eg. 2100, Chess.com gets lower.

+2000 = every extra 50 lichess points, difference increases by 25.

eg. 2100 lichess = 1850 chess com.

Through this difference, we can also say —

1950 chess.com = 2150 Lichess = 1900 FIDE.

"1900 FIDE is your estimated score as measured with the same amount of your chess.com games."

The reason for the relationship between lichess and chesscom ratings is because of the rate of change for wins and losses. On this site, you are awarded 8 points for a win against an opponent of the same rating, on lichess you are only awarded 6 points. If you put it on a graph, you'd see that you could predict exactly at what level a player's expected rating on chesscom would surpass their expected lichess rating. Fide awards ~5 points for a win against a similarly rated opponent so I would expect that you would need a 2000 rapid rating on chesscom to be similar to a 1900 fide rating.

But they will win and lose similar amounts. Say plus six for a win and minus six for a loss. It does mean that a Chess.com rating is more volatile than Lichess or FIDE and therefore prone to greater inaccuracy but that inaccuracy can be either way .... up or down.

That only matters given a player maintains a 50% win rate. Say, a player maintains a win rate of 60% against similarly rated opposition. This means over the course of 100 games they will be +20 to the win column. that equates to 100 points fide, 120 points lichess and 160 points on chesscom (not exactly as they would get less rating as their rating goes up and/or their win rate would go down as they faced higher rated opponents but the basic idea stands). Basically, until a player reaches their plateau and hits a 50% win rate, they will settle at a higher rating on chesscom the stronger they are. Fide rating will be the lowest. Lichess is in the middle. The difference in ratings would be more clear cut if each used the same initial rating.

Silent_Assassin

These responses you are posting — Fide rewards at least 20 for a nearby rating, and for the highest rating 39.8 per win. It makes sense as you only play either 7, 9 or 11 games per tournament of standard time control. Of course, in a month could be more than that, but I'm sure it's never more than how much you could play here. The calculation I gave is absolutely correct, passing through the starting ratings of all 4 chess playing organizers/platforms.

Forums
Forum Legend
Following
New Comments
Locked Topic
Pinned Topic