I think it is perfect
Between 326-360 points difference would be my first guess
Let compare the best blitz rating of some known GM here
Just for fun
Lets see GM Magnus Carlsen
Lichess Blitz is 3165
Fide Blitz is 2986
3165 - 2986 = 179 points difference
Let see GM Nihal Sarin
Lichess Blitz 2995
Fide Blitz : 2685
310 points of difference
Lets see GM Alireza Firouzja
Lichess Blitz : 2856
Fide Blitz : 2810
46 points of difference
Lets see GM Shakhriyar Mamedyarov
Lichess Blitz : 2923
Fide Blitz : 2866
67 points of difference
On 4 players the average rating difference is 150.5 points
I'm surprised how close the blitz rating is between Lichess & Fide !
Let compare the best blitz rating of some known GM here
Just for fun
Lets see GM Magnus Carlsen
Lichess Blitz is 3165
Fide Blitz is 2986
3165 - 2986 = 179 points difference
Let see GM Nihal Sarin
Lichess Blitz 2995
Fide Blitz : 2685
310 points of difference
Lets see GM Alireza Firouzja
Lichess Blitz : 2856
Fide Blitz : 2810
46 points of difference
Lets see GM Shakhriyar Mamedyarov
Lichess Blitz : 2923
Fide Blitz : 2866
67 points of difference
On 4 players the average rating difference is 150.5 points
I'm surprised how close the blitz rating is between Lichess & Fide !
@kasparovbab said in #22:
> Between 326-360 points difference would be my first guess
>
> Let compare the best blitz rating of some known GM here
> Just for fun
>
> Lets see GM Magnus Carlsen
>
> Lichess Blitz is 3165
> Fide Blitz is 2986
>
> 3165 - 2986 = 179 points difference
>
> Let see GM Nihal Sarin
>
> Lichess Blitz 2995
> Fide Blitz : 2685
> 310 points of difference
>
> Lets see GM Alireza Firouzja
>
> Lichess Blitz : 2856
> Fide Blitz : 2810
> 46 points of difference
>
> Lets see GM Shakhriyar Mamedyarov
>
> Lichess Blitz : 2923
> Fide Blitz : 2866
> 67 points of difference
>
> On 4 players the average rating difference is 150.5 points
>
> I'm surprised how close the blitz rating is between Lichess & Fide !
No sir, you never extrapolate what you find on the extremes and apply it to the medium.
I mean, you can measure how much deviation they have, but it is going to be very different from the medium. This happens in any topic.
The percentage of aggression between man and woman is around 55-45, 60-40 something lika that, but if you go to the extreme and find the top 100 most aggresive persons, 98 of them are men.
The average person exercises like 1- 2 hours a day. Bodybuilders exercise like 5-8 hours a day.
The average calorie intake for men should be around 2500. but obese and bodybuilders consume double or triple.
The average chess player wins and loses at about the same rate their games. The top players win around 25-40% more than they lose.
The average salary in the US is around 57,000. The extremes make millions.
Just dont do that.
> Between 326-360 points difference would be my first guess
>
> Let compare the best blitz rating of some known GM here
> Just for fun
>
> Lets see GM Magnus Carlsen
>
> Lichess Blitz is 3165
> Fide Blitz is 2986
>
> 3165 - 2986 = 179 points difference
>
> Let see GM Nihal Sarin
>
> Lichess Blitz 2995
> Fide Blitz : 2685
> 310 points of difference
>
> Lets see GM Alireza Firouzja
>
> Lichess Blitz : 2856
> Fide Blitz : 2810
> 46 points of difference
>
> Lets see GM Shakhriyar Mamedyarov
>
> Lichess Blitz : 2923
> Fide Blitz : 2866
> 67 points of difference
>
> On 4 players the average rating difference is 150.5 points
>
> I'm surprised how close the blitz rating is between Lichess & Fide !
No sir, you never extrapolate what you find on the extremes and apply it to the medium.
I mean, you can measure how much deviation they have, but it is going to be very different from the medium. This happens in any topic.
The percentage of aggression between man and woman is around 55-45, 60-40 something lika that, but if you go to the extreme and find the top 100 most aggresive persons, 98 of them are men.
The average person exercises like 1- 2 hours a day. Bodybuilders exercise like 5-8 hours a day.
The average calorie intake for men should be around 2500. but obese and bodybuilders consume double or triple.
The average chess player wins and loses at about the same rate their games. The top players win around 25-40% more than they lose.
The average salary in the US is around 57,000. The extremes make millions.
Just dont do that.
The way to create a justification would be to find out your predicted rating, then go get a rating, and report back on its accuracy.
Until a lot of people do that it has no statistical justification. @ChessMathNerd said in #18:
> @Molurus Thank you very much for the detailed response! That sort of information is exactly what I was looking for.
> The reason I am trying to figure this out is because I am trying to reach FM, and I want to know what kind of rating I need on lichess before entering FIDE is worth my time.
That's why I vote down every single person who posts data like this. It causes so much harm. Here you are, claiming to want to be good at OTB chess, and not doing it based on a couple graphs, that have no mathematical validity at all.
Here's the real issue. The people who have the will to go do it, and try to improve, are not like the people who think of excuses not to do it.
You can't derive your predicted success from people who are putting in effort and practice at something you're not willing to do.
Until a lot of people do that it has no statistical justification. @ChessMathNerd said in #18:
> @Molurus Thank you very much for the detailed response! That sort of information is exactly what I was looking for.
> The reason I am trying to figure this out is because I am trying to reach FM, and I want to know what kind of rating I need on lichess before entering FIDE is worth my time.
That's why I vote down every single person who posts data like this. It causes so much harm. Here you are, claiming to want to be good at OTB chess, and not doing it based on a couple graphs, that have no mathematical validity at all.
Here's the real issue. The people who have the will to go do it, and try to improve, are not like the people who think of excuses not to do it.
You can't derive your predicted success from people who are putting in effort and practice at something you're not willing to do.
Yes I will continue to do that !!! I'm comparing the same things sir @Alientcp
It's all about the highest blitz rating of each players.
For @Sarg0n according to my calculation his highest fide blitz rating would be 2228.5
And I'm doing an ''approximation'' of the highest fide blitz rating
lets see what google said ... about ap·prox·i·ma·tion
Learn to pronounce
noun: approximation; plural noun: approximations
a value or quantity that is nearly but not exactly correct.
"these figures are only approximations"
h
Similar:
estimate
estimation
guess
conjecture
rough calculation
rough idea
surmise
guesswork
guesstimate
ballpark figure
semblance
outward appearance
likeness
resemblance
similarity
correspondence
comparison
a thing that is similar to something else, but is not exactly the same.
"the band smashed up their equipment in an approximation of rock star behavior"
It's all about the highest blitz rating of each players.
For @Sarg0n according to my calculation his highest fide blitz rating would be 2228.5
And I'm doing an ''approximation'' of the highest fide blitz rating
lets see what google said ... about ap·prox·i·ma·tion
Learn to pronounce
noun: approximation; plural noun: approximations
a value or quantity that is nearly but not exactly correct.
"these figures are only approximations"
h
Similar:
estimate
estimation
guess
conjecture
rough calculation
rough idea
surmise
guesswork
guesstimate
ballpark figure
semblance
outward appearance
likeness
resemblance
similarity
correspondence
comparison
a thing that is similar to something else, but is not exactly the same.
"the band smashed up their equipment in an approximation of rock star behavior"
@kasparovbab said in #26:
> lets see what google said ... about ap·prox·i·ma·tion
You can call it however you want, but it doesnt mean its accurate. It may be for them in the extremes, but it will be totally off for the medium if you extrapolate. I mean, you can do whatever you want, i really dont care, im just telling you that the extrapolation of those numbers to the medium would be as accurate as getting a number out of a hat, or a RNG from a webpage.
> lets see what google said ... about ap·prox·i·ma·tion
You can call it however you want, but it doesnt mean its accurate. It may be for them in the extremes, but it will be totally off for the medium if you extrapolate. I mean, you can do whatever you want, i really dont care, im just telling you that the extrapolation of those numbers to the medium would be as accurate as getting a number out of a hat, or a RNG from a webpage.
@Alientcp You too you can do whatever you want with your numbers ... I just prefer my approximation to the one given in this website ethanlebowitz.github.io/RatingConverter/index.html
Im not giving a number. i noticed that Sarg0n is lower in that site than he should be, as a CM he had to be 2200 FIDE at least at 1 point, im just assuming he has maintained the level as a minimum, so im skeptical of the number the site returns.
But if you compare the blitz rating of the best chess players in both fide and lichess , you only get an approximate of the top extreme of a graph that we do not have. It doesnt work on the medium.
Again, the average chess player has like a 47 W% 47 L% and about 6% draw rate.
The top 15-20 GMs are measured different.
ratings.fide.com/profile/1503014/statistics
Carlsen 396 wins, 309 draws, 79 lost. 50%+ wins 39%+ draw. 10% lost.
It will be similar for the elite. So, when you extrapolate the same averages the elite have to the medium, your ruler wont work either, at least not with the average player. The average player wins as much as he loses. The elite rarely lose. (and the worst rarely win)
Im just telling you that dont use the same ruler you use on the extremes on a graph to measure the average, it doesnt work, both extremes, high and low deviate too much to be a reliable ruler for the actual average.
I mean, you can measure them, there may be a correlation, just dont extrapolate to the average and expect to be accurate.
But if you compare the blitz rating of the best chess players in both fide and lichess , you only get an approximate of the top extreme of a graph that we do not have. It doesnt work on the medium.
Again, the average chess player has like a 47 W% 47 L% and about 6% draw rate.
The top 15-20 GMs are measured different.
ratings.fide.com/profile/1503014/statistics
Carlsen 396 wins, 309 draws, 79 lost. 50%+ wins 39%+ draw. 10% lost.
It will be similar for the elite. So, when you extrapolate the same averages the elite have to the medium, your ruler wont work either, at least not with the average player. The average player wins as much as he loses. The elite rarely lose. (and the worst rarely win)
Im just telling you that dont use the same ruler you use on the extremes on a graph to measure the average, it doesnt work, both extremes, high and low deviate too much to be a reliable ruler for the actual average.
I mean, you can measure them, there may be a correlation, just dont extrapolate to the average and expect to be accurate.
This topic has been archived and can no longer be replied to.