Congratulations to johnmusacha (#984) for necroing this. So much intelligent discussion has occured since.
Merci, although a fortnight between comments 983 and 84 isn't really "necroing".
What about three weeks between #1002 and #1003 ?
On a side note, I always wonder how necromancers find threads. The ones they posted in, I understand, but when they are just posting for the first time some "I agree, lol" ?
What about the three weeks? Does this bother you?
Don't you have anything else to worry about other than what goes on, in the Chess.com forums?
Do you think you are some kind of gendarme? Or the Forum Surete?
Sorry if I offended your tender sensibilities. Perhaps you might find other forums more to your liking, sir.
Its clear that based on the facts provided by the folks that didn't come here just to argue and scream "apples and oranges hurr durr" for two years, that generally speaking a person's Chess.com blitz ratings would be about 100-300 points lower than that same person's USCF rating.
So it's 100-300 now?
Both Chess.com and USCF are based on Glicko systems.
Almost. Chess.com uses the glicko rating system, but the USCF uses a modified elo system. Now, glicko is, in some sense, based on elo, so they are related, but they use very differet systems. In fact, it is the differences in glicko and elo that lead to deflation.
Glicko is an elaboration on elo. It adds a second statistic called RD. You can consider elo a special case of glicko where all RDs = 0, in fact.
However, in glicko, all RDs are not 0. And it is these differences in RD between two players that imbalances the number of points won/lost in a game. When a low RD player beats a high RD player (which actually happens slightly more often, because lower RD players are more experienced), the loser loses more points than the winner wins. Since this happens more often, over time the total number of rating points of all players will go down (simplification, since the number of players will change, but you can think of the average rating going down if you prefer). This is basically what is meant by "deflation."
I wrote Professor Glickman this, when asking him for descriptions of ratings:
"I would imagine there are a few key cognitive skills/knowledge related to chess skill:
I don't understand post #1010, can someone please explain it?
He's saying that since chess.com has such a huge ratings discrepancy when compared with USCF, then the ratings here should be adjusted. I agree.
Glicksman did not address why one rating system would be different from another. It would appear that wasn't the question that was asked.
Glicksman also did not refute that Chess.com blitz ratings are deflated against USCF ratings, either. As an expert in the field, Glicksman definitely would not have accepted the premise of said deflation if he did not believe it to be true.
To wit, he would have responded "What do you mean the ratings are deflated against USCF? That's absolute idiocy! Apples! Oranges!"
So, wait the gentleman that invented the Glicko rating system says that the chess.com ratings need to be adjusted?
That's pretty significant in itself.
I did not see where he said that.
I did not see where he said that.
You must have some severe reading comprehension problems then, bubba.
We should just go back to the OLD Russian standard: are you beginner or master? Nothing else.
I did not see where he said that.
You must have some severe reading comprehension problems then, bubba.
If I'm not mistaken, and I don't think I am, the part you think Glickman said was said by the author of the post. I find it unlikely Glickman would say anything like chess.com's ratings need "adjusting."
Nope you got the two parts switched up in your mind brah
I did not see where he said that.
You must have some severe reading comprehension problems then, bubba.
If I'm not mistaken, and I don't think I am, the part you think Glickman said was said by the author of the post. I find it unlikely Glickman would say anything like chess.com's ratings need "adjusting."
But Glickman himself did say that, and it's clearly written in the email. Don't shoot the messenger, brah. I'm just passing on the good news.
Comparing the two rating systems is pointless - they are completely different mathematical algorithms that are formulated in such a manner that no meaningful data can be extracted from the comparison. Hence any approximations made of the difference between the two rating systems are going to be invalid and hence no valid conclusion can be made. End of story. Source: statistics degree
Funny how professor Glickman said nothing of the sort when he was asked about it. I think I'd trust the inventor of the Glicko system over your "statistics degree". Thanks for trying though, brohans.
Its clear that based on the facts provided by the folks that didn't come here just to argue and scream "apples and oranges hurr durr" for two years, that generally speaking a person's Chess.com blitz ratings would be about 100-300 points lower than that same person's USCF rating.