Ratings

[MostFamousDane]

The newest ratings have been posted: http://wcdf.wz.cz/ratings.htm

[Alex_Moiseyev]

I don't understand how Alex can be ranked 8.

Easy - wrong math model. I am very happy with such effect. When I was ranked "only" #2 - noone wanted to listen my arguments about weaknesses of Glico rating system. Maybe now they will look more seriously at this paradox. I wasn't very successful in the beginning of this year finishing second in TN State ty by winning only 4 games from 16 with no losses. In IL state ty in April It was also not a very perfect performance. However dropping almost 100 points is absurd.

I hope Eric Strange read this and now he should very small doubts - what rating system and formula must be used for ACF ratings.

I already preliminary calculated: If I win my WCM against Ron with 4-5 points margin, my rating rank will jump from #8 to #3 or #4.

[MostFamousDane]

I don't understand how Alex can be ranked 8.

Because he had a worse performance than the 7 players above him. I always find it funny that people complain when the ratings come out - expecting them to reflect their own subjective opinon on who is the best players. That is not what ratings are meant for - ratings reflect objectivly the relative PERFORMANCE of the players.

[MostFamousDane]

When I was ranked "only" #2 - noone wanted to listen my arguments about weaknesses of Glico rating system.

I don't remember the discussion but if your starting point was that "ok I'm the greatest player in the world and I'm only rated nr 2 something must be wrong with the system", I can undestand why people didn't want to listen for the reason I mentioned above. If the argument was "I won every single game in the last period and my rating still dropped" or something similar, then the only reason I can think of is that we are all out to get you - muaaaaahahaha, oh wait am I not supposed to say that .

Anyway cheer up at least you only dropped 7 places I dropped 60

[Alex_Moiseyev]

Anyway cheer up at least you only dropped 7 places I dropped 60

This system works against active players - as more you play, as more you lose. I don't complain - I am laughing.

I always thought that on long run any proper rating system should rank players, not performance. This is the way how ELO works. But WCDF is unique organisation and has unique rating system

Maybe Igor can explain something. I am dummy with all this logic and don't understand it at all.

AM

[tommyc]

I think everyone knows the more you play the more your position in the ratings drop,isnt that why certain top players only play occasionally and then usually in soft Tys?? Sure its as plain as th nose on your face.

[william]

Hi everyone ,

I just came back from looking at the ratings update , and although I am "" happy"" with my 2 place progression , in looking further I ask myself , should it have been more ?? should I have dropped 10 instead of 2 up.

I will only take 1 example ... Joe davidson ( a guy I really like and appreciate) should never be in top 10 and I am sure that if he could then he would say this himself , as there are so many top quality players below him!! in ratings .

I believe that although a ratings system is needed , and needs time to stabilize , it must have a coherent and solid base otherwise anything and everything is possible , and there seems to be a little something missing here.

I still encourage though the poor guys who have to tear their hair out to get this info out to the checker playing world , but would ask that the system be reworked , so that it is fair to ( as alex says) the players who play in the tough tys and not just the mickey mouse ones.

As far as a world champ being classed no.1 in ratings this would seem logical except that we have not only 3 move but gayp also . So I would say that it doesn't really matter as world champion title is considered higher that no.1 anyway and any intelligent player will admit that about 10 players in the world right now are capable of hitting 1st spot...

As far as ratings are concerned I have already stated that one should concentrate on serious study and the ratings will take care of themselves whether it be a bad or good system ,if you're good ; you'll go up!! and as far as I am concerned the only thing that counts is getting my world 3 move title , so 4th or 40th ...who cares

best wishes

WILLIAM

[tommyc]

No chance Billy ,without "Cannings Compilations." hehehe........

[shinzen]

Anyway cheer up at least you only dropped 7 places I dropped 60

This system works against active players - as more you play, as more you lose. I don't complain - I am laughing.

I always thought that on long run any proper rating system should rank players, not performance. This is the way how ELO works. But WCDF is unique organisation and has unique rating system

Maybe Igor can explain something. I am dummy with all this logic and don't understand it at all.

AM

Yes Alex, Igor it can consecrate. Was you are player with same high rating and 'you had on every tournament high engagement. Never you are him dishonour, so that you are fell. I calculate your (you and ron) probable rating, whether would you are play today and you won from 40 games 5 games and 35 games draws. Alex 2567, Ron 2429
Regards,
Igor
p.s. GLicko is perfect system for active players.
p.s.2 every years same tears

[Alex_Moiseyev]

you won from 40 games 5 games and 35 games draws. Alex 2567, Ron 2429

I accept your promises However, even if this happened, my rating will start dropping again when I play in State tournaments. This already happened in the past. In some situations I have to win 10-12 games out of 14 in order to support my rating. There is no way under the sun to do this by playing 6-8 tournaments a year.

Again, I don't worry much about this list, but other people may have questions. Rating cannot be dropped so significantly from 1-2 "bad" events, where I didn't win almost all games.

All I can do - promise my fans and ACF members that my "local-state" activity will not be dropped, or lets say - rating never will be the reason of such drop.

[shinzen]

Alex, I agree with You (and I am your fan too ). But You are mathematic and know, what is statistic and statistic method. ELO and GLICKO are statistic method. Ratings is thereof, who is most successful in tournament and no, who is best players. Best player is world champion promiscuous ratings place
Greetings,
Igor

[Eric Strange]

I like how everyone is trying to defend the Glicko System. The fact is that I have been studying both Glicko and ELO and Glicko system is complete BS... The WCDF should really consider getting with the times and upgrading their math model.

The Glicko system is completely against Active players and support inactive players who play once every 3 years.

The fact is.... The Glicko system is discouraging players to be active tournament players which is the opposite of what needs to be done.

The Glicko system was good before the days of the internet when players got their serious challenging play at tournaments. Players who got challenging play 10 times a years at a tournament vs someone who barely got to play the game but once every 3 years. The player with less time played got rewarded.

The fact is that with the internet in place players can sit on the computer and play challenging game after game after game. Players who don't go to a tournament but once every 3 years may play EVERYDAY online vs players just as good as top masters (and they get to play them 100 games in a row daily).
Igor... I would like for you to send me an email, eric.j.strange@gmail.com, so we can discuss rating a little bit. I have taken over ACF ratings and would like to hear some of your views.

Thanks you,

Eric Strange

[Alex_Moiseyev]

Igor... I would like for you to send me an email, eric.j.strange@gmail.com, so we can discuss rating a little bit. I have taken over ACF ratings and would like to hear some of your views.

It would be great if both of you guys communicate and exchange experience / ideas.

Igor, Eric will be doing ACF rating processing, and currently he is in the middle of very important step - ANALYZING and making decision - what system to use for ACF rating.

Eric, Igor is very nice, friendly professional guy and player, and GREAT FAN of GLICKO system

Igor, the Russian quote says that "all politicians and statistics are for sale !" Here is one of my favor statistical research, which proves dangerity of smoking:

All smoking people who was born before 1870, passed away !

Regards,

Alex

[shinzen]

[quote="Alex_MoiseyevIt would be great if both of you guys communicate and exchange experience / ideas.

Igor, Eric will be doing ACF rating processing, and currently he is in the middle of very important step - ANALYZING and making decision - what system to use for ACF rating.

Eric, Igor is very nice, friendly professional guy and player, and GREAT FAN of GLICKO system

Igor, the Russian quote says that "all politicians and statistics are for sale !" Here is one of my favor statistical research, which proves dangerity of smoking:

All smoking people who was born before 1870, passed away !

Regards,

Alex[/quote]
Alex, You are small mistake . I am not Glicko fan. But glicko is for historical reason. I'm just a follower . I like clasical ELO (Czech Union of Draughts rate ELO system only). Transfer from the GLicko to ELo is easy and quickly, but problem are with players from B list. WCDF Committee need change rating rules (how much games need to A list, how long time (2,3-5 years, etc.). You understand me , I am loyal to currently committee , must contact committee
Greetings,
Igor
p.s. Eric me e-mail is ikeder@volny.cz

[Alex_Moiseyev]

WCDF Committee need change rating rules (how much games need to A list, how long time (2,3-5 years, etc.). ... I am loyal to currently committee

Thanks, Igor !

Who is on WCDF Rating Committee and should there a motion procedure being initiated ?

I suggested to Alan Millhone and he agreed with me to use "5 years limit" for ACF rating "B" list. In addition, I think 50 games must be used for "A" list for new players and 25 - for players who move from list "B" to list "A".

50 games in average means 4-5 events, 25 games - 2-3 events.