Checking if Elo system is oppressive [With proofs]

Sort:
Avatar of MasterJyanM

another game of which i won,... just sharing because of amazeing comeback

at least i'm not a suck-up like jankowski

i play as white

Avatar of TitanMaster101

hes just stating facts

Avatar of magipi

Amazing that this absurd topic is still alive after more than a month. Although it seems to slowly evolve into another Janko troll show.

Avatar of BigChessplayer665
magipi wrote:

Amazing that this absurd topic is still alive after more than a month. Although it seems to slowly evolve into another Janko troll show.

Cause the owner is a janko troll alt it was always a janko show

Avatar of basketstorm
magipi wrote:

Amazing that this absurd topic is still alive after more than a month. Although it seems to slowly evolve into another Janko troll show.

What's absurd about the topic?

Avatar of IeJoker

Yeah that's definitely not ChatGPT

Avatar of magipi
basketstorm wrote:
magipi wrote:

Amazing that this absurd topic is still alive after more than a month. Although it seems to slowly evolve into another Janko troll show.

What's absurd about the topic?

We already discussed this, haven't we? What the whole thing is based on is not a real statistical analysis, but instead a fake one made by a BS generator, ChatGPT.

Avatar of basketstorm

Sure but I already answered and will answer again since you are repeating this lie: you are terribly wrong about my usage of ChatGPT. You think I've used language model to get numbers - wrong, instead ChatGPT just helped with the code and all numbers are result of actual calculations. No fakes.

Avatar of magipi
basketstorm wrote:

Sure but I already answered and will answer again since you are repeating this lie: you are terribly wrong about my usage of ChatGPT. You think I've used language model to get numbers - wrong, instead ChatGPT just helped with the code and all numbers are result of actual calculations. No fakes.

Obviously we're not going to convince each other. What I say is: you used a text generator for something that it isn't good for, the results you got were absurd, so those results are certainly wrong. You say that they're fine, everyone else is wrong, all those statisticians and mathematicians who developed and tested the Elo-system many decades ago were all wrong.

Avatar of basketstorm

Read my other posts there's so much more issues than what's revealed in the simulation from original post. Don't focus on that, that was just a starter. Arpad Elo created the Elo system, is he wrong? No. Read his book like I did. And you'll realize how flawed chess.com ratings pool/system is.

Avatar of RandomChessPlayer62
basketstorm wrote:

Sure but I already answered and will answer again since you are repeating this lie: you are terribly wrong about my usage of ChatGPT. You think I've used language model to get numbers - wrong, instead ChatGPT just helped with the code and all numbers are result of actual calculations. No fakes.

"Well, I have a semi-coherent nonsense generator in my garage, it doesn't know how to code but knows the general structure of a piece of code. Don't say it is incorrect, it didn't give me numbers, it gave me code. All the numbers were from real calculations."

See how bad that sounds when describing ChatGPT as what it is when I argue it back at you?

Avatar of basketstorm
RandomChessPlayer62 wrote:
basketstorm wrote:

Sure but I already answered and will answer again since you are repeating this lie: you are terribly wrong about my usage of ChatGPT. You think I've used language model to get numbers - wrong, instead ChatGPT just helped with the code and all numbers are result of actual calculations. No fakes.

"Well, I have a semi-coherent nonsense generator in my garage, it doesn't know how to code but knows the general structure of a piece of code. Don't say it is incorrect, it didn't give me numbers, it gave me code. All the numbers were from real calculations."

See how bad that sounds when describing ChatGPT as what it is when I argue it back at you?

Not sure what are you trying to say. I went through every line of that code, ChatGPT did exactly what I asked. I could've written that myself, ChatGPT just saved me some time. Now, read #868

Avatar of ElijahTheAnew

I'm 500 so this makes me feel better. If I lose - they were always 1000+ rated tongue

Avatar of Andriisarchus

im not educated enough to understand this

Avatar of basketstorm
XPGK99 wrote:
Andriisarchus wrote:

im not educated enough to understand this

Basically, their point is everyone is so good at chess that it's hard for anyone to move up to the elo they should be. (I might be wrong but that's what I inferred.)

Point is rating system here is flawed due to multiple factors, including ones Arpad Elo (creator of Elo ratings) warned about. Ratings here are very inaccurate, both inflated and deflated and with serious localized distortions. This contributes to a frustrating experience of a wildly inconsistent strength of opponents within same rating range.

Avatar of TitanMaster101

I THOUGHT WE WOULD LET THIS ROT PEOPLE

Avatar of vshahistocoolforu

let it rot tomato.

Avatar of vshahistocoolforu

stop trying to start the FIDE revolution

Avatar of vshahistocoolforu

never heard of u. stop the Fédération Internationale des Echecs révolution.

Avatar of vshahistocoolforu

its not bullying. if i have never heard of u until now, then I can state that without consequences. also stop trying to overthrow chess.