The "Daily rating" rebalancing change has been reverted...
Sort:
So what occurs within the bell curve - when extra points are added solely to 1500 and above?
The curve becomes even more top heavy. The higher the rating = the greater separation. Rating inflation is increased at the top end. Those below remain stuck. Especially true by adding 400/ 300 points.
Nikki- I stated your formula was incorrect... as 40M active accounts is an assumption. The discussion revolves about just what the Stats are representative of. Your figures are in the ball park - assuming total accounts = active players on a daily basis. I’m not convinced this is case.
The stats suggest my 1600+ rapid rating is in the top 3%. Which appears to represent that of every 100 games being played - 3 are of players above my rating while 96 games are between lower rated players. I just don’t think this is the case for live play pairing pools. The figures may be there within the grand scope of all players. But the majority simply are not playing. They have a number assigned to the account and that’s about it. It was stated the Stats were thought to be of players completing 20 games. I think a better stat would be to view how many players are available in any given pool.
Jan 23, 2021
0
#66
40M was the most recent figure that I read. If you Google “how many users are there on chess.com” that’s the figure it spits out. I’m sure it’s more now.
And yes, that was a ballpark figure. There was rounding to simplify the math, but this doesn’t change the narrative, just the actual numbers. What difference would it make if it were 3.6M or 2.9M? The answer is none.
And the formula is not incorrect. It is absolutely the correct way to calculate the size of a pool given the information at hand.
The number of games played is irrelevant to calculating pool size given the data. Once the parameters of what constitutes the pool are set, the calculation follows simply.
You can argue that those parameters should be set differently, which is fine. I don’t care, I really don’t. The number of couplings I give couldn’t be smaller. I just feel a need to defend myself when I’m laughed at and told I know not of my math, especially since the answers are so obvious. I made no errors aside from rounding for the sake of simplicity.
And yes, that was a ballpark figure. There was rounding to simplify the math, but this doesn’t change the narrative, just the actual numbers. What difference would it make if it were 3.6M or 2.9M? The answer is none.
And the formula is not incorrect. It is absolutely the correct way to calculate the size of a pool given the information at hand.
The number of games played is irrelevant to calculating pool size given the data. Once the parameters of what constitutes the pool are set, the calculation follows simply.
You can argue that those parameters should be set differently, which is fine. I don’t care, I really don’t. The number of couplings I give couldn’t be smaller. I just feel a need to defend myself when I’m laughed at and told I know not of my math, especially since the answers are so obvious. I made no errors aside from rounding for the sake of simplicity.
Sorry that you feel attacked Nikki. Simply not the case. I disagreed with one of your inserted variables - as being a valid number and had nothing to do with any mathematical calculation.
Formulas, graphs , bell curves all represent a truth when viewed sans data. What gets inserted becomes key. The numbers chosen along the baseline or the time involved along the vertical line can represent most anything the operator chooses.
Jan 23, 2021
0
#69
BadBishopJones3 wrote:
The stats suggest my 1600+ rapid rating is in the top 3%. Which appears to represent that of every 100 games being played - 3 are of players above my rating while 96 games are between lower rated players. I just don’t think this is the case for live play pairing pools. The figures may be there within the grand scope of all players. But the majority simply are not playing. They have a number assigned to the account and that’s about it. It was stated the Stats were thought to be of players completing 20 games. I think a better stat would be to view how many players are available in any given pool.
It gives the number of active players in the pool, where activity is at least 20 games over the last 90 days in that pool. One can disagree if that definition is sufficient, but it's what the site chose.
But other than the minimum number of games, the percentile has nothing to do with the number of games or the pairings you're going to get in any particular session. That's driven by what the player ratings are when you are on and what seeks match your seeks at that time.
Youtube - VM FernandesLuiz chess
Chess.com - https://www.chess.com/member/fernandesluiz
Twitch - https://www.twitch.tv/vmfernandesluiz
Here’s a pool of players
Here’s your ranking within a pool of players.
This is the number ahead of you.
Calculating the pool size is trivial.
You are the one claiming that this must be wrong somehow. You are making the claim that goes against the evidence. Therefore it is up to you to provide counter-evidence undermining the evidenced claim.
You’re simply making complicating assertions that boil down to “this can’t be right.”
I mean I’m sure it’s possible that there is some flaw in the chess.com algorithm, but I don’t see how that could be possible except for perhaps some kind of typo. What possible motive would they have to purposefully mislead us?
Honestly, I don’t care one way or another. I was just trying to do the math to help settle a dispute and I get mocked! Jeez. People sometimes!