Your network blocks the Lichess assets!

lichess.org
Donate

Introducing a universal rating converter for 2024

I wish I was 2200 FIDE as the converter says :(

I wish I was 2200 FIDE as the converter says :(

hello
my rating in rapid is 1500 on lichess
I have a classical rating tho
my highest rating was 1535
how does this match
??

hello my rating in rapid is 1500 on lichess I have a classical rating tho my highest rating was 1535 how does this match ??

@AshwinChunchu you are 2200 lichess classical. According to the converter, that puts you close to 1900 FIDE. Blitz and bullet are not considered.

@AshwinChunchu you are 2200 lichess classical. According to the converter, that puts you close to 1900 FIDE. Blitz and bullet are not considered.

@MBurns2020 Because you have all of your profiles linked together, I was able to explore you specifically. Your lichess classical = 2030, chesscom rapid = 2100, and FIDE = 1900. Based on my converter, I would guess that both your chesscom rapid and FIDE seem accurate as they are quite close together, and you are underrated on your lichess classical rating. Indeed, when I look at your FIDE plot, you've gained a lot of rating this past year (200+ FIDE in one month! Nice job), and yet you haven't played a lichess classical game in months.

So I would disagree with your assessment actually. I would say your are almost perfectly modeled by my converter. Play some lichess classical games and I would guess you could make it to 2200 in no time!

@MBurns2020 Because you have all of your profiles linked together, I was able to explore you specifically. Your lichess classical = 2030, chesscom rapid = 2100, and FIDE = 1900. Based on my converter, I would guess that both your chesscom rapid and FIDE seem accurate as they are quite close together, and you are underrated on your lichess classical rating. Indeed, when I look at your FIDE plot, you've gained a lot of rating this past year (200+ FIDE in one month! Nice job), and yet you haven't played a lichess classical game in months. So I would disagree with your assessment actually. I would say your are almost perfectly modeled by my converter. Play some lichess classical games and I would guess you could make it to 2200 in no time!

@RogerH3 Actually I did this modeling using Julia. As a mathematician, I didn't need any packages (beyond basic math and linear algebra functions that Julia already provides) to derive best fit lines, compute r values, etc. Basic stat functionality is not that complicated really.

@RogerH3 Actually I did this modeling using Julia. As a mathematician, I didn't need any packages (beyond basic math and linear algebra functions that Julia already provides) to derive best fit lines, compute r values, etc. Basic stat functionality is not that complicated really.

I love this!
Finally I can compare my OTB DWZ with my online rating and also convert it to an expected FIDE ELO. Great job!

I love this! Finally I can compare my OTB DWZ with my online rating and also convert it to an expected FIDE ELO. Great job!

You forgot to convert score on lichess to elo points.

Also make a table wich responds to ultrabullet.

Ultra is real chess.

Play not good, play smart!

:3

You forgot to convert score on lichess to elo points. Also make a table wich responds to ultrabullet. Ultra is real chess. Play not good, play smart! :3

This blog is completely wrong Im FIDE 1600(I play OTB very often) and Chesscom 1950 and this guy is saying Chess.com 1950 is 1800 FIDE

This blog is completely wrong Im FIDE 1600(I play OTB very often) and Chesscom 1950 and this guy is saying Chess.com 1950 is 1800 FIDE

I enjoyed reading the blog authors various replices as added informative bits about this whole world of conversions, but also the rating systems practical implementation traditions or problems with regard to the mutliplicity of systems and their pool dependencies (and then the events that define the pools of games underlyying the clound of rating values to make the most coherent ensemble).

That extra data world reasoning arguments have to be leveraged in the multiplicty of families of function ambient space restrictions that one would then find optimal values.

But it does seem that none of the thinking has been considering the more robust machine learning view, that although was obligatory for NN ambient function spaces to explore for various objective of global optmization, could also be an alternate angle to gage the generaliztaion robustness of each of the various extra-data consideration based various regressions experiement I have hear clues about in such conversation (I might have skipped some chunks, I always do so I could be corrected by linking back to previous posts, that would be nice). I based my reading more on blog authors well articulated replies.

I am also more of a tourist here. Not ever having had experience of clear picture of all the non-lichess rating systems and their evolution and their implementation in the field changes.. etc... But this kind of post blog discussion, is like a gold mine to find hint and crumbs of the reasoning patterns that might have been invovled in such world of ratings systems applied to chess.

I enjoyed reading the blog authors various replices as added informative bits about this whole world of conversions, but also the rating systems practical implementation traditions or problems with regard to the mutliplicity of systems and their pool dependencies (and then the events that define the pools of games underlyying the clound of rating values to make the most coherent ensemble). That extra data world reasoning arguments have to be leveraged in the multiplicty of families of function ambient space restrictions that one would then find optimal values. But it does seem that none of the thinking has been considering the more robust machine learning view, that although was obligatory for NN ambient function spaces to explore for various objective of global optmization, could also be an alternate angle to gage the generaliztaion robustness of each of the various extra-data consideration based various regressions experiement I have hear clues about in such conversation (I might have skipped some chunks, I always do so I could be corrected by linking back to previous posts, that would be nice). I based my reading more on blog authors well articulated replies. I am also more of a tourist here. Not ever having had experience of clear picture of all the non-lichess rating systems and their evolution and their implementation in the field changes.. etc... But this kind of post blog discussion, is like a gold mine to find hint and crumbs of the reasoning patterns that might have been invovled in such world of ratings systems applied to chess.

@BuzzardChecker said in #68:

What we know is that FIDE decided their ratings were misleading, but they did not really get to the bottom of why that happened. One must assume the problems will emerge slowly again
IMHO the biggest problem here is that FIDE is aware - or at least it experts are - that the design of the ELO rating system is the actual problem and any cosmetic tweaks will only work as a temporary mitigation. But reading between the lines, FIDE seems to be afraid that a more modern rating system also means a more complex and therefore harder to understand for general public. (Plus some political aspects resulting in things like the questionable "400 points rule".) Just see how many lichess users have trouble understanding how rating deviation works and how it affects rating updates. With all its deficiencies, ELO has one big advantage: anyone is able to follow the calculations and verify easily that their rating updates are correct if they feel an urge to do so. With something like Glicko-2 or other complex rating system, it wouldn't be as easy and that might harm the trust in the rating system.

@BuzzardChecker said in #68: > What we know is that FIDE decided their ratings were misleading, but they did not really get to the bottom of why that happened. One must assume the problems will emerge slowly again IMHO the biggest problem here is that FIDE is aware - or at least it experts are - that the design of the ELO rating system is the actual problem and any cosmetic tweaks will only work as a temporary mitigation. But reading between the lines, FIDE seems to be afraid that a more modern rating system also means a more complex and therefore harder to understand for general public. (Plus some political aspects resulting in things like the questionable "400 points rule".) Just see how many lichess users have trouble understanding how rating deviation works and how it affects rating updates. With all its deficiencies, ELO has one big advantage: anyone is able to follow the calculations and verify easily that their rating updates are correct if they feel an urge to do so. With something like Glicko-2 or other complex rating system, it wouldn't be as easy and that might harm the trust in the rating system.