Chess dot com vs FIDE ratings

Sort:
Oldest
Lit

As a patzer who sucks at chess I was wondering if I am FIDE underrated?

I know chess.com and Fide use different rating calculation methods called Glicko and Elo respectively and that they have different player pools but yet there must be a correlation between the two.

I'm 1900 here and only 1300 FIDE, is this much difference normal? I don't drop below 1800 on chess.com even when I'm tilted and also on my alt account I can sustain 1800 easily. But beating even 1000 Elo rated players in Fide tournaments is never easy for me.

jasonarthurtaylor

Some people are really good at shorter games like bullet but when it comes to classical time formats they do less well. Others, like myself, are the opposite. So, there are three ELOs for any one person. 
FIDE tournaments are more likely to be classical time formats.
There's a known equation relating USCF and FIDE, USCF = 180 + 0.94*FIDE.
I know chess.com is constantly adjusting their ELOs. They actually try to avoid matching up FIDE stuff on time formats. That's why the slowest time formats here are not called classical but are called rapid instead, is my theory.

Lit
jasonarthurtaylor wrote:

Some people are really good at shorter games like bullet but when it comes to classical time formats they do less well. Others, like myself, are the opposite. So, there are three ELOs for any one person. 
FIDE tournaments are more likely to be classical time formats.
There's a known equation relating USCF and FIDE, USCF = 180 + 0.94*FIDE.
I know chess.com is constantly adjusting their ELOs. They actually try to avoid matching up FIDE stuff on time formats. That's why the slowest time formats here are not called classical but are called rapid instead, is my theory.

Thanks for answering. I have to add that I am comparing my chess.com Blitz with my FIDE Blitz. FIDE also has Blitz, but theydon't have bullet. Real pieces will fall down on the ground if you play ۱ min game on real board.

blueemu

This site is grotesquely over-rated compared to FIDE, but there is no direct conversion factor that applies to all ratings. A player at one end of the rating band might be 300 points over-rated, while a player at the other end of the band might be rated 500 points too high.

satan_llama

Blitz - 300 points= FIDE (IMO). It also depends on the regional strength. Though 1300 is really low imo ( I am myself 1100 lol coz I don't play much). If you play more tournaments you should be able to get to 1500.

Lit

Today they updated (increased) FIDE ratings for players rated below 2000. New rating = Old rating + (2000 - old rating) * 0.4 . What do you think? Why did they do this? Now this site's rating is closer to FiDe rating ig

medelpad
Lit wrote:

Today they updated (increased) FIDE ratings for players rated below 2000. New rating = Old rating + (2000 - old rating) * 0.4 . What do you think? Why did they do this? Now this site's rating is closer to FiDe rating ig

I like it, it makes my rating higher

Uhohspaghettio1

It just goes to show how unreliable and arbitrary FIDE ratings are. Online ratings are undoubtedly a much better indicator of strength. You have a nice bullet rating.

eathealthyfoods

I don't know...

tygxc

@7

"Today they updated (increased) FIDE ratings for players rated below 2000. New rating = Old rating + (2000 - old rating) * 0.4" ++ FIDE also raised the rating floor to 1400.
FIDE did this because there was a problem with many underrated people in the lower ratings.

"Now this site's rating is closer to FiDe rating ig"
++ Still chess.com rapid rating = FIDE rapid rating + 100

@11

"It just goes to show how unreliable and arbitrary FIDE ratings are."
++ On the contrary, the FIDE rating is reliable and steady, chess.com is overrated by about 100.

"You have a nice bullet rating"
++ Bullet is another game. FIDE has classical rating, rapid rating, and blitz rating, but no bullet.

tygxc

@14

That will not happen.
To calibrate chess.com ratings to FIDE ratings everybody would lose rating on chess.com.

blueemu
Uhohspaghettio1 wrote:

It just goes to show how unreliable and arbitrary FIDE ratings are. Online ratings are undoubtedly a much better indicator of strength. You have a nice bullet rating.

Can't figure out whether this is sarcasm or wishful thinking...

eathealthyfoods
blueemu wrote:
Uhohspaghettio1 wrote:

It just goes to show how unreliable and arbitrary FIDE ratings are. Online ratings are undoubtedly a much better indicator of strength. You have a nice bullet rating.

Can't figure out whether this is sarcasm or wishful thinking...

I don't know either...

tygxc

@19

Chess.com is still overrated relative to FIDE: about +100 in rapid and +400 in blitz.

Uhohspaghettio1
tygxc wrote:

@19

Chess.com is still overrated relative to FIDE: about +100 in rapid and +400 in blitz.

It has already been explained to you in the other thread this only applies to the highest levels, the elite players. It has been a thing for many years that the rapid ratings here are a lot easier to get at the lower levels here than blitz. As you don't seem to believe me and would rather make up your own numbers out of your imagination here's a source:

https://chessgoals.com/rating-comparison/#chesscomotb

This has been found by multiple people on multiple surveys over multiple years. If you can work out the slightly clunky website and you look at 2000 chess.com blitz that person has an average of 1915 FIDE plus or minus 80 and an average chess.com rapid of also 2000 plus or minus 110.

This survey doesn't actually show the FIDE ratings actually overtaking at really low levels like had been found in the past. Besides this is certainly the case now after the recent massive boost.

There is something weird about your manner and how you engage with the forums. You constantly and repeatedly make statements that are either false or go against the common traditional advice. For example telling someone that any person studying openings is a complete waste of time when they asked for opening advice on that board. And then you don't give any response to getting called out, you don't apologize or make any sort of argument in your favour, just continue stating the same thing the next time.

Stop giving misinformation, people are coming here to get good information and not garbage.

tygxc

@20

"this only applies to the highest levels"
++ No, this applies to all levels if ratings are stabilised.
It falls short for players massively underrated because in the phase of steep progress.
Proof:
Probability of Carlsen beating Player X in rapid on chess.com
= probability of Carslen beating Player X in rapid on FIDE.
Hence
Chess.com rapid rating Carlsen - chess.com rapid rating Player X
= FIDE rapid rating Carlsen - FIDE rapid rating Player X
Hence
Chess.com rapid rating Player X - FIDE rapid rating Player X
= chess.com rapid rating Carlsen - FIDE rapid rating Carlsen
= 100
Also
Chess.com blitz rating player X - FIDE blitz rating player X 
= chess.com blitz rating Carlsen - FIDE blitz rating Carlsen
= 400

"either false" ++ No.

"go against the common traditional advice" ++ Yes, when common traditional advice is false.

"studying openings is a complete waste of time" ++ It is. Capablanca said the same.
The opening has no influence on the outcome of a lower level or fast time control game.
What you study does not happen and when it finally happens you will have forgotten.
It is much better to find your own moves on the board than reeling off memorised moves, as at some point you have no more memorised moves and you have to find your own moves on the board anyway.

"make any sort of argument in your favour" ++ I do.

"people are coming here to get good information" ++ I give plenty and for free.

medelpad
tygxc wrote:

@19

Chess.com is still overrated relative to FIDE: about +100 in rapid and +400 in blitz.

Idk how that works, I'm 1900 online in both blitz and rapid

Uhohspaghettio1
11mik wrote:
medelpad wrote:
tygxc wrote:

@19

Chess.com is still overrated relative to FIDE: about +100 in rapid and +400 in blitz.

Idk how that works, I'm 1900 online in both blitz and rapid

That’s interesting. Usually the player pool in blitz is much stronger than the one in rapid. So I’d say if you are 1900 blitz you just need to play a bit more rapid to easily get a higher elo, maybe around 2100 rapid. That said, probably the player pools even out at higher elos. (Of course I’m talking about 10min rapid and 3min blitz or 3/2. If you play 5/5 blitz maybe it’s possible to have a higher blitz rating, because the strongest blitz pool is at those said faster time controls.)

Right. It evens out at around the level of 1900-2100, which is ironically what tygxc's rating level is now. Thousands of people answering surveys have confirmed it. Those people have established rapid and blitz ratings, they're not going to answer a survey knowing that their rapid rating is not accurate. Besides, the plus/minus are literally right there, it's in that confidence interval.

It doesn't work like your "proof" tygxc. Many years ago they stated that chess.com was doing a change to their rating system and it wasn't using the original elo formula anymore, they even have a different formula for lower levels and higher levels. Many things can happen in rating systems.

Even if we look at your "proof", the highest rated rapid player on chess.com is an IM named Sean Winshand and his rapid rating is 2892. No super GM is busting their balls and getting outplayed by a no-named IM, they are not taking it with much seriousness. According to you he should be 3192 blitz, essentially being the third highest blitz rated on the site after Hikaru and Carlsen. He should also have a 2792 OTB rating according to you, putting him also number 3 in the world behind Carlsen and Caruana.

https://www.chess.com/leaderboard

His blitz rating is 2956, so even at the very top his blitz rating is just 64 points higher than his rapid.

Isn't it funny how you don't play blitz at all here yourself and yet are making these assertions? Isn't that funny? And it universally agreed by everyone else how blitz are tougher until higher levels where it starts to even out. No doubt your own blitz rating would be 2400+ right?

So here's a challenge - either play blitz yourself here and prove it or stop giving this misinformation. If you like you can go to custom times, more times and change it to 5 5 or 5 2, though you might need to wait slightly longer for a game. It'll be fairly close to 10 0 as you play - a 40 move 5 5 game is 8 minutes and 20 seconds, almost 10 minutes and will go higher if you need more time for more moves. Increment will also help avoid time rushes. Meanwhile you likely typically end the game significantly before 10 0 like you play with a bit in the tank. In other words it's literally the same time limit as what you play except fairer.

So play blitz and let's see you get to 2400, according to you your blitz rating should be 2410 or thereabouts. It won't take long to see you are around the same as your rapid in blitz if not lower. Now maybe it will be 2200 or something. According to the website, 2105 rapid should be 2200 in blitz on average, plus or minus 110. So you will be 2090-2310 blitz if you play period. But you claim 2410 according to your current rating, you're going to be beating 2500s regularly right? Will be nice to see you beating 2500s and the occasional 2600 where lots of current active IMs battle it out with your zero openings and your Evan's Gambit.

Keep in mind you yourself are at moderate levels, at the lower levels it gets progressively that the rapid rating is higher than blitz.

Furthermore...

""studying openings is a complete waste of time" ++ It is. Capablanca said the same."

That is a total lie, Capablanca never said that.

"The opening has no influence on the outcome of a lower level or fast time control game."

Again, this is totally untrue it as was shown and proved in how every blitz tournament at high levels make heavy use of openings. When the evaluation bar fluctuates wildly over the course of a game, only then do openings not matter.

"What you study does not happen and when it finally happens you will have forgotten."

Saying such things don't make them true.

"It is much better to find your own moves on the board than reeling off memorised moves, as at some point you have no more memorised moves and you have to find your own moves on the board anyway."

Noone recommends that. Show me an opening book that gives you long lists of variations, advises you to learn them by heart and says you're on your own then. The whole point of studying openings is not just exact moves other than the very first ones, but the way the game unfolds. It's in some ways just another method of studying chess, regardless of the actual moves.

tygxc

@24

"chess.com was doing a change to their rating system and it wasn't using the original elo formula" ++ Chess.com uses Glicko-2,
which is an improved version of elo: converges faster but requires more calculations.

"they even have a different formula for lower levels and higher levels" ++ No. elo has 3 different K-factors and assumes the same constant rating deviation RD for all; Glicko-2 calculates a personalised rating deviation RD and K-factor for each player at all times.

"According to you he should be 3192 blitz" ++ Not according to me.
Classical, rapid, blitz, and bullet are different games with differet skills and different ratings.

"Isn't it funny how you don't play blitz"
++ 'I play way too much blitz chess. It rots the brain just as surely as alcohol.' - GM Nigel Short

"making these assertions?" ++ Mathematical assertions

"Capablanca never said that." ++ He did.
'Ninety percent of the book variations have no great value, because either they contain mistakes or they are based on fallacious assumptions; just forget about the openings and spend all that time on the endings.' - Capablanca

Also: 'Memorization of variations could be even worse than playing in a tournament without looking in the books at all.' - Botvinnik

"When the evaluation bar fluctuates wildly over the course of a game, only then do openings not matter." ++ In lower level games and fast time controls the evaluation bar fluctuates wildly.

"What you study does not happen and when it finally happens you will have forgotten."
++ It is true. E.g. a player studies Alekhine's Defense, but all of his opponents play 2 Nc3. When after months some plays 2 e5 the player will have forgotten. E.g. a player studies the Marshall Attack, but all his opponents avoid it with 8 a4 or 8 h3 or 8 d3. When after months one plays 8 c3 the player will have forgotten. Even Caruana admits he often has trouble recalling his own prepared lines.

"It is much better to find your own moves on the board than reeling off memorised moves, as at some point you have no more memorised moves and you have to find your own moves on the board anyway."
++ recommended e.g. by Dorfman.

GYG
Uhohspaghettio1 wrote:

Stop giving misinformation, people are coming here to get good information and not garbage.

This is a suitable reply to >99% of what tygxc posts. He is probably the most consistently incorrect person I've ever seen on chess.com.

Forums
Forum Legend
Following
New Comments
Locked Topic
Pinned Topic