great question, im wondering the same thing but the stupid site owner dont wanna answer that question nor any other members =(
How accurate is the percentile ranking really?
Sort:
Jan 28, 2016
0
#3
Let's not be too hard on chess.com, guys. They run a great site, and think a little about that stat question:
Where would you draw the line between an "active player" and one who isn't active enough to be counted?
# of games/month?
# of days/weeks/months with no games?
Minimum # of games/week/month/year?
I think if someone is willing to pay for a membership for more than one year, then he/she should be counted.
*Maybe* there could be some kind of minimum activity for a first year member (after their first 10 games they get counted?), but after that, I think every renewing member should count.
KISS if you know what I mean.. :)
Jan 28, 2016
0
#4
Imagine if it said that there were 831,863 and you were ranked 831,863. Some one must hold this exhalted position...and you remained there for several weeks. But the only way is up or would you give up.
I'm curious as to how chess.com calculates percentile rankings for its users. Right now when I go onto my stats, under live standard play, it shows I'm ranked 180,863 of 831,863 players. This shows as a percentile of 78.3%. This is correct.
What I'm asking though is if this percentile really tells me anything? Is chess.com including only active players in the ranking and percentile calculation (meaning users who show regular activity), or does the 831 863 players include everyone who's ever made an account, and not closed there account if they decide to discontinue use of the site?
If the latter is the case then I would argue that the calculated percentile is not very useful, as the value would be highly skewed upwards. I have to imagine a lot of people make an account, play a few games, and decide not to make chess a hobby, leaving themselves with a low rank that stays in the system.
If chess.com accounts for this somehow that would great.