It was very befuddling to me, though, I am probably in a state of fuddlement already.
Chess.com Proposal: How to Handle Growing Bullet Ratings
Sort:
I mean that I'm befuddled by it - a lot; Like, really befuddled.
I've got it! He means your own rating has to be within the seek range you set. I have no idea how that helps but I'm amazed that I've gotten this far and really I think I should receive a trophy or something for this.
I'd settle for a little shadenfreude at those bullet carbunklers. Everybody pick door number 2!
I feel all woozy now.
I like option 3.
Just simply take away 200 from everyone.
May 20, 2011
0
#27
waffllemaster wrote:
Fezzik wrote:
From a purely mathematical sense, the chopping method is certainly the cleanest method.
Still, from a human perspective, a gradual reduction will probably be worth the headaches.
My thoughts too.
Couldn't disagree more, a slow bleed will be an indicator to players that their playing strength is falling, which will cause angst and frustration. It's also likely to cause some players to not play as much due to negative feedback.
Chopping everybody's rating feels more drastic initially, but it's a one time thing that's much easier to explain in a single note. It's also much more transparent and will apply to everybody equally.
I prefer the third option .
I would chop them. My highest rating is bullet, which annoys me.
I don't really need the ratings to be in line with those of national fedrations. I understand that ratings are a relative measure of stregth within a certain rating pool-no more, no less. If ratings here were 100-1200, yet still accurate within their rating pool, I wouldn't mind.
And I second the thought to stop giving out the 1200 rating to unrateds. An adult beginner is much lower, an average club player much higher. Just have them play a certain amount of games before establishing a rating.
@ TheMouse : Your idea is totally commercial and shortsighted, its just like to suggest that a school lowers its standards and let everyone succeed even if they dont deserve so,just for the sake of getting more students that will pay the fees. Dont you see the strategic impact of such a decision? it would actually lower the site's credibility especially among the good players and chess world in itself and after sometime even the ego boosted players would eventually leave it because it wont attract the attention they seek (as a great site with quality).
May 20, 2011
0
#32
themouse wrote: I would leave bullet ratings as they are, and gradually raise blitz and standard ratings 150 points
Raise standard ratings here 150 pts??? I'm 2063 here (standard); 1869 (ASPCC) and 1539 (IECC) and guess which rating is the most inflated?
ANSWER: my chess.com rating
I've also noticed that the blitz ratings here are grotesquely inflated as well. Why not just start everyone off at 2800 to "even things out?"
May 20, 2011
0
#33
LordNazgul wrote:
Puchiko wrote:
I don't really need the ratings to be in line with those of national fedrations. I understand that ratings are a relative measure of stregth within a certain rating pool-no more, no less. If ratings here were 100-1200, yet still accurate within their rating pool, I wouldn't mind.
I think that it is very desirable that the ratings be in line with the ratings of the established federations as much as possible. Otherwise the players who don't play in OTB tournaments (who are probably the majority) don't have an accurate measure of their strength / progress.
That isn't even possible. The players in each pool are different. The number of games in each type of online play are different than the number of games that can be played in OTB tournaments (can't play 20, 30, 40+ games at a time OTB and be rated). Plus, in this particular case, there isn't a comparable rating class in OTB play.
Ratings only need to be accurate within a particular pool of players at a given time control. All they are is a statistical model on how one player would fare against another player; i.e. the likelyhood of one player winning over the other.
May 20, 2011
0
#34
Well, you do make a very good point here Erik, and yes i agree that something needs to be done, so option 1 i would disagree with likewise.
Option 2 i belive would caus ea lot of problems and confusion, slowly bleeding raings, means that members realise there ratings are dropping slowly, and some dont notice, those who dont they notice later on and later and later and so on, so you will be getting loads of complaints over a long period of time and it is disorientating for the members.
OPTION3 to me seems the best, It s done all at once, its easy to send a message to everyone about 200 being cut and new formula implementation.
Any complaints will be short lived and will not be staggered over a liong period of time.
Just my opinion anyway :D
Regards
EchaniX13
I still think we should go with option 2 and deny all responsibility.
May 20, 2011
0
#36
ivandh wrote:
I still think we should go with option 2 and deny all responsibility.
Except that there is this entire thread on it :), not the most dicreet way and if lying to members is ever found out, that brings down the reputation big time, and is very hard to get back :)
LordNazgul wrote:
Puchiko wrote:
I don't really need the ratings to be in line with those of national fedrations. I understand that ratings are a relative measure of stregth within a certain rating pool-no more, no less. If ratings here were 100-1200, yet still accurate within their rating pool, I wouldn't mind.
I think that it is very desirable that the ratings be in line with the ratings of the established federations as much as possible. Otherwise the players who don't play in OTB tournaments (who are probably the majority) don't have an accurate measure of their strength / progress.
Further, I'd contend that the bullet ratings aren't even accurate within the pool.
What makes players unable to track their progress is a sudden "adjustment" of ratings by 200 points. If the rating is accurate within the rating pool, to track your strength, all you need to do is compare past and present rating.
However, it's true that non-OTB players would probably like to know how strong they are in OTB terms. And even though that's impossible to state precisely, a chess.com rating could be somewhat an estimate. On the other hand, others might argue that comparing a chess.com blitz/correspondence to an OTB 2*90 min makes about as much sense as comparing snooker and scrabble ratings. The conditions are different, the rating pool is different, the ratings will be different.
However, I think trying to adjust chess.com to OTB ratings only decreases their accuracy within the rating pool. Whether that's a justifiable price to pay is up for discussion. For me, it isn't. However, I believe it is for the majority.
I would divide the bullet rating group into two new groups:
1. pure 1 minute games (i.e. 1|0 tempo)
2. other bullet games (i.e. 1|1, 1|2, 2|0, 2|1 tempos)
I think that bullet ratings are overrated for players who specialize mostly on 1|0 bullet games. But why would you chop 200 rating points from players who play mostly 2|1 bullet games?
You can argue this point back and forth but the simplest way to do this, without any long term problems is to make the quick cut. Gradual reductions with formula changes will lead to mathematical headaches. We can go back and forth about inflated ratings but does this bring us any closer to a solution? Of course there is a difference in true ratings when comparing regular length chess games to blitz games. There's also a difference between online ratings and face to face ratings. I think the point here is make the most painless alteration in ratings points in the easiest manner.
cdir wrote:
waffllemaster wrote:
Fezzik wrote:
From a purely mathematical sense, the chopping method is certainly the cleanest method.
Still, from a human perspective, a gradual reduction will probably be worth the headaches.
My thoughts too.
Couldn't disagree more, a slow bleed will be an indicator to players that their playing strength is falling, which will cause angst and frustration. It's also likely to cause some players to not play as much due to negative feedback.
Chopping everybody's rating feels more drastic initially, but it's a one time thing that's much easier to explain in a single note. It's also much more transparent and will apply to everybody equally.
I kinda considered that everyone would be aware of what's going on... but I see what you mean a lot of players don't check their messages or post on the forum so I guess many would take it as negative feedback. #3 seems good to me now 
How about this: the seek must include one's own rating. (As in turn-based) This would prevent people from collecting point by point from much weaker opponents on industrial basis, as well as those who lurk waiting to get a shot at a high-rated opponent and run with the points.
Okay, I can't understand this. I'm a quitter