Does chess.com have the toughest pool of players?
Sort:
Oct 25, 2017
0
#2
Or perhaps the rating system on the other site isnt the same as on chess.com?
Hmm that is a possibility. Actually speaking of that is chess.com's glicko rating nearly the same as the one used by FIDE
Oct 25, 2017
0
#4
i know what site you're talking about but think about it. if everyone starts at 1500 on the other site then all the newish players (800-1000 here) will deflate to around 1200 1300 while the beginners (around 1000-1200/1300 here) who are the majority will hover around 1500 and the intermediate players (1300/1400-1600) will hover around 1600 or 1700. the advanced players (1600-2000) will hover around 1800 to 2200.
although what i say isn't 100% accurate my point still stands that you will get more elo points on the other site lichess than here.
I'm also higher at rapid on "the other site" than I am at daily here. Something's going on.
Oct 25, 2017
0
#6
Rating is always relative to the pool of players. That should answer your question.
Oct 25, 2017
0
#8
Rating is not an absolute number. All it does is let you compare to another player in the same pool.
An Elo-based rating system can be set up with any mean. It could be 1200 (chess.com is roughly around here), it could be 1500 (the other site might be here), it could be zero. There's no mathematical reason why it even has to be a positive number, although that would be a bit unconventional.
What you should look at is not the actual number by your name, but what percentile you're in.
Oct 25, 2017
0
#9
klimski wrote:
Rating is always relative to the pool of players. That should answer your question.
More specifically, relative to the subset of the rating pool that you've actually played. Unless you research every opponent's history carefully, you don't really know how many games they have under their rating belt (more games = more accurate rating) or what level of opposition they faced.
Until you've played a few dozen games at each time control, your rating is subject to huge variation based on a large number of factors beyond overall "site strength."
Oct 25, 2017
0
#10
Brontide88 wrote:
klimski wrote:
Rating is always relative to the pool of players. That should answer your question.
More specifically, relative to the subset of the rating pool that you've actually played. Unless you research every opponent's history carefully, you don't really know how many games they have under their rating belt (more games = more accurate rating) or what level of opposition they faced.
Until you've played a few dozen games at each time control, your rating is subject to huge variation based on a large number of factors beyond overall "site strength."
Once you've played a few dozen games against a variety of opponents at or near your strength, your rating will be relatively accurate for the rating pool you are in. You don't need to play everyone else in the rating pool to be able to make a meaningful comparison to others in the same rating pool.
Oct 25, 2017
0
#11
Brontide88 wrote:
klimski wrote:
Rating is always relative to the pool of players. That should answer your question.
More specifically, relative to the subset of the rating pool that you've actually played.
And also the subset you've played is subject to the subset they've played, etc etc. Since most people play random opponents though, it's not a big issue.
Brontide88 wrote:
Until you've played a few dozen games at each time control, your rating is subject to huge variation based on a large number of factors beyond overall "site strength."
Since most people stick to 1 or 2 time controls per rating type (e.g. only 3|0 and 5|0 in blitz) if you do this then your rating may be less reliable because your results are coming from different groups.
Interestingly you may be able to achieve a similar effect by playing during different hours of the day since chess.com members wouldn't be equally distributed across all time zones.
Actually the percentile system is a good measuring point as someone pointed out above. Mgleason most likely.
On chess.com I am just short of 85% (84.change) for both rapid and blitz. I would like to get better obviously but touching 85% is not the worst.
On the other site, I dont know if there is even a way to check, will take a look though.
Oct 25, 2017
0
#13
JayeshSinhaChess wrote:
Actually the percentile system is a good measuring point as someone pointed out above. Mgleason most likely.
On chess.com I am just short of 85% (84.change) for both rapid and blitz. I would like to get better obviously but touching 85% is not the worst.
On the other site, I dont know if there is even a way to check, will take a look though.
assuming [a particular site] then you have to click your name in the top right corner of the screen and then click on the time control you want to look at and it should say 'GLICKO2 RATING (your rating) Better than (a number) percent of (time control for example blitz) players.
you've obviously never played on ICC
Okay I checked and this is what it said for blitz -
Glicko2 rating: 1479.4. Better than 44.1% of Blitz players.
Progression over the last twelve games: 8. Rating deviation: 63.95.
That is stunningly low. I mean 85% on chess.com and 44% on the other site. Wow
Oct 25, 2017
0
#16
saying you're better than 44% of blitz players is the same as saying you're in the 44% percentile. (i think) which means that most people on that site have a higher rating. if you were in the 44% percentile here it would mean that most people are better than you so it's the same thing.
Oct 25, 2017
0
#17
although it's quite difficult to get below the 50% percentile on this website because most of the 19 million accounts on here aren't active and for the majority of them they have their ratings abandoned at a really low rating. i'm not sure i'm correct in my reasoning but i know i'm correct in the fact it's hard to be in a low percentile.
Oct 25, 2017
0
#18
Yeah, I'm not sure to what extent abandoned/inactive accounts are included in the reported percentiles. If the two sites report it differently, you might not be able to make a meaningful comparison based on that.
It might also be interesting to compare the gaps between your rating and whoever's at the top of the leaderboard.
yes...100% besides bullet
Chess.com don't have stronger players
It's just chess.com starts you off at 300 elo lower then most sites
When I First played on Chess.com they started you at 1500 and my elo was pretty much identical to all other chess sites
Tbh I believe chess.com out of all the sites has the bigger share of cheaters
Only thing better about chess.com is the training content .... If you willing to Pay
I recently joined another website, sorry chess.com, I am cheating on you, and there I found that the elo I have is around 200 points higher than what I have here. Atleast for blitz I have not played classic there a lot.
I won't name the website but its not just any website, it is also quite popular and feature a wide variety of games. So I figure that the crowd here on chess.com must be really good. I can't defeat a 1500 elo player here, while on that other site wins and defeats are equal number, but there I am 1500 plus change on blitz. On chess.com I touched 1400 once only to slip down again.
So is this correct, that chess.com has the best group of players.