Page 1 of 2 12 LastLast
Results 1 to 10 of 13

Thread: Preliminary Policy on Online CFC rated games

  1. #1
    Join Date
    Jan 2009
    Location
    Tecumseh, ON
    Posts
    3,268
    Blog Entries
    1

    Default Preliminary Policy on Online CFC rated games

    The CFC is continuing to work on details of Online Chess regulations. Especially troubling are the allegations of cheating. The following is our current policy.

    Any player found guilty of violating a site's Fair-Play policy during a CFC Rated tournament is subject to the following standard suspension from CFC rated online play.
    Age under 12 = 4 months

    Age under 16 = 8 months
    Others = 12 months
    Some organizers may have a Fair-Play Committee consisting of strong players to assess games in place of or in addition to the above (relying on a site's policy). Any players found to have received outside assistance by such a Committee will also be subjected to the same penalties.


    Any player who has their account closed for a fair-play violation will not be able to play in CFC tournaments in the future on that site, by simply creating a new account, without permission of the site.
    There is an Appeals process. The appeals fee is $100 to be paid to the CFC Business Office, and will be refunded if the appeal is successful. Exact details will be determined case by case but will involve the game(s) being re-evaluated using anti-cheating software and expert opinion. The player may be asked to play supervised games to evaluate his ability, and assess his claims of innocence. Any additional costs involved in the appeal will be the responsibility of the player and are non-refundable.

  2. #2
    Join Date
    Aug 2008
    Location
    Kanata, Ottawa, Ontario
    Posts
    1,227

    Default

    good job CFC, it can't be easy navigating these new rapids

  3. #3

    Default

    This policy is very dangerous and could expose the CFC to a lawsuit.

    Being called a cheater online does not have the same level of seriousness as being called a cheater by the CFC. An online server standard of proof could be lower that the FIDE standard that the CFC is expected to follow.

    I was a member of the committee that catch the first convicted cheater in Quebec and I cannot tell you his name because under Quebec Laws, Truth is not an absolute defence against a libel civil case. We must prove that there is a public interest in the publication of the Cheater's name and according to the FQÉ attorney, no such public interest exist. Only FQÉ approved Organisers and Arbiters are allowed to know the name of the Cheater.

    Use of strong players is a poor idea. We should use FIDE approved experts like Dr. Regan from University of Buffalo who is quite close to us. A strong player opinion is not a statistical analysis. FIDE uses statistical analysis as part of the anti-cheating process. Beginning with the null hypothesis that the player his honest, the final product is a Z score that express the level of confidence of the null hypothesis. Suspending anybody who didn't plead guilty without a statistical proof is high risk.

    I will not post too much details on the investigation process because knowing how its done could help cheaters to avoid prosecution.

    ARB and FPL did held a common seminar on the fight against cheating https://fpl.fide.com/seminars/ . The seminar introduce very important details: for example, scanning device (metal detectors) are mandatory at all FIDE rated events. The metal detector cannot be hidden in a closet, it must be clearly visible to the players in order to act as a deterrent. Failure to comply may result in a refusal to rate the tournament. The CFC has the Authority to issue waivers to some anti cheating requirements.

    FPL considers that anti-cheating has little to do with arbitration and is considering the creation of the anti-cheating specialist title considering that persons other than arbiters could be better qualified for anti-cheating duties.

    Anti-cheating, in standarplay chess and in online chess is going to take more of our resources in the years to come. It would be worthwhile to have two elected commissions to take care of cheating: one that decided whether or not to sue and another that judge that case. FIDE did emphasized the fact the the two functions: making the accusation and judging the case, should be carried-out but two separate entities.

    Quote Originally Posted by Anti-cheating protection measures
    Tournament organizers shall adopt one of the three levels of the AC Protection Measures: standard protection, increased protection, or maximum protection. These levels of protection are to correspond with the three types of tournaments identified in Section

    1. Compliance with the AC Prevention Measures is mandatory. All organizers shall inform FIDE Qualification Commission (QC) and/or their National Federation that they are complying with the AC Prevention Measures, by so stating in the Tournament Application form.

    Upon request from a tournament organizer and based on good cause, the National Federation and/or Fide QC may grant a waiver from some of the requirements enumerated below to tournaments with Standard or Increased levels of protection. The waiver request must be submitted in advance and describe adequate anti-cheating measures that are tailored to the tournament’s size and budget. The organizers are not limited on the number of waivers they can submit.

    Tournaments that are found not to materially comply with the AC Protection Measures shall not be rated.

    In Tournaments designated for increased or maximum levels of protection, a special Anti-Cheating Arbiter may be employed. An Anti-Cheating arbiter is an arbiter who is entrusted with anti-cheating tasks (and who may have received special anti-cheating training) and must be equipped with at least one hand held scanner. The employment of the Anti-Cheating Arbiter is mandatory in tournaments that require maximum level of protections.

    Organizers of events designated for increased and maximum protection are strongly encouraged to consult with the ACC on finding adequate AC measures that are tailored to the tournament’s size and budget. ACC’s decision following contact by the organizers and subsequent consultation with WCOC and/or any other relevant Fide Commission is final.

  4. #4
    Join Date
    Jan 2009
    Location
    Tecumseh, ON
    Posts
    3,268
    Blog Entries
    1

    Default

    Dr. Regan has been consulted on at least one of the cases of suspected cheating. His software came to same conclusion that I did in that case. We have not publicly named anyone. I believe that there are only two cases outstanding.

  5. #5
    Join Date
    Jan 2009
    Location
    Tecumseh, ON
    Posts
    3,268
    Blog Entries
    1

    Default

    Quote Originally Posted by Pierre Dénommée View Post
    This policy is very dangerous and could expose the CFC to a lawsuit.

    Being called a cheater online does not have the same level of seriousness as being called a cheater by the CFC. An online server standard of proof could be lower that the FIDE standard that the CFC is expected to follow.

    I was a member of the committee that catch the first convicted cheater in Quebec and I cannot tell you his name because under Quebec Laws, Truth is not an absolute defence against a libel civil case. We must prove that there is a public interest in the publication of the Cheater's name and according to the FQÉ attorney, no such public interest exist. Only FQÉ approved Organisers and Arbiters are allowed to know the name of the Cheater.
    Thank you Pierre for pointing this out to all the readers. I was aware of this anomaly in Quebec libel law but there may be others who are not.

    With regard to online servers and playing in online CFC events the following guidelines are helpful (which I already published elsewhere in another thread).

    1. Do not change the focus away from the game window during rated games. Do not open the Word document with your shopping list or manifesto for world domination during a rated game.
    2. Do not have chess engines or programs like chessbase or forwardchess which might open an engine running in the background.
    3. Do not get assistance during your game from any source.

    If you follow those rules, you will likely not get into trouble with the online chess platforms.

    The CFC is only able to address allegations of cheating online that occur during CFC rated play. We have no interest in regulating all online play or even have the capacity to deal with all the different online chess platforms and disputes that might arise. If you cheat in CFC play and this cheating is detected, there will be consequences. Organizers of CFC events take cheating very seriously as you are threatening the very integrity of the game. I know of one organizer that has contacted Dr. Regan directly and had him run at least one game through his analysis software. We have his email. We will not be shy to use it.

    If you sue the CFC we will sue you back unless you sue us for some trivial amount. We will fight all lawsuits as long as I am president. The cost of such a lawsuit will not be trivial unless you can get a lawyer for free. We cannot sit back and allow people to cheat as that would be the end of online chess.
    Last edited by Vladimir Drkulec; 08-11-2020 at 03:37 AM.

  6. #6
    Join Date
    May 2009
    Posts
    48

    Default

    This report by Alex Holowczak is a very good read:

    http://www.4ncl.co.uk/data/Article_on_Fair_Play.pdf

  7. #7

    Default

    Quote Originally Posted by Vladimir Drkulec View Post
    Thank you Pierre for pointing this out to all the readers. I was aware of this anomaly in Quebec libel law but there may be others who are not.

    With regard to online servers and playing in online CFC events the following guidelines are helpful (which I already published elsewhere in another thread).

    1. Do not change the focus away from the game window during rated games. Do not open the Word document with your shopping list or manifesto for world domination during a rated game.
    2. Do not have chess engines or programs like chessbase or forwardchess which might open an engine running in the background.
    3. Do not get assistance during your game from any source.

    If you follow those rules, you will likely not get into trouble with the online chess platforms.

    The CFC is only able to address allegations of cheating online that occur during CFC rated play. We have no interest in regulating all online play or even have the capacity to deal with all the different online chess platforms and disputes that might arise. If you cheat in CFC play and this cheating is detected, there will be consequences. Organizers of CFC events take cheating very seriously as you threatening the very integrity of the game. I know of one organizer that has contacted Dr. Regan directly and had him run at least one game through his analysis software. We have his email. We will not be shy to use it.

    If you sue the CFC we will sue you back unless you sue us for some trivial amount. We will fight all lawsuits as long as I am president. The cost of such a lawsuit will not be trivial unless you can get a lawyer for free. We cannot sit back and allow people to cheat as that would be the end of online chess.
    Rules 1, 2 and 3 are excellent.

    Cheating must indeed be taken seriously, this is obvious from the joint seminar and has been obvious since FPL existed. On the one hand, we should never back down because of a lawsuit, but on the other hand, we must clearly follow all the steps required to get a strong case against the player. It is the solidity of our proof combined with the respect of the proper procedures that will discourage the suspended player from suing.

  8. #8
    Join Date
    Aug 2008
    Location
    Kanata, Ottawa, Ontario
    Posts
    1,227

    Default

    So I already saw that report, and I'm trying to wrap my ahead around something: is the ROI calculation not dependent on the initial rating submitted?

    Thanks Stephen (et al)

  9. #9
    Join Date
    May 2009
    Posts
    48

    Default

    I'm no expert in such matters and I don't know about the ROI per se, but Regan's entire method is based on ratings:

    "Ken Regan's algorithm is based on the player's rating - the idea being that a 1700 playing 2500 moves over a prolonged period is cheating, whereas a 2500 playing 2500 moves isn't.

    The strength of your opponents isn't relevant, because playing 2500 moves is the same regardless of your opponent. But your own rating is fundamentally relevant, because it's the bar you're being measured against."

    [This taken from the English Chess Forum, where there have been two long threads (50+ pages) on fair play.]

    Thus having an accurate input rating is important, which can be an issue when players have differing national and FIDE ratings, for example, or the underrated improving junior.
    Last edited by Stephen Wright; 08-12-2020 at 04:59 PM.

  10. #10
    Join Date
    Aug 2008
    Location
    Kanata, Ottawa, Ontario
    Posts
    1,227

    Default

    Quote Originally Posted by Stephen Wright View Post
    I'm no expert in such matters and I don't know about the ROI per se, but Regan's entire method is based on ratings:

    "Ken Regan's algorithm is based on the player's rating - the idea being that a 1700 playing 2500 moves over a prolonged period is cheating, whereas a 2500 playing 2500 moves isn't.

    The strength of your opponents isn't relevant, because playing 2500 moves is the same regardless of your opponent. But your own rating is fundamentally relevant, because it's the bar you're being measured against."

    [This taken from the English Chess Forum, where there have been two long threads (50+ pages) on fair play.]

    Thus having an accurate input rating is important, which can be an issue when players have differing national and FIDE ratings, for example, or the underrated improving junior.
    Thanks Stephen. My concern is especially the underrated improving junior. I have a gnawing suspicion, that given the inherent higher skill variability demonstrated by lower-rated players, that such statistical methods are better at catching cheaters at the 2000 level than they are at the 1400 level.

    Please note though that I have zero math to back that up, this is just perceptual.

Page 1 of 2 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •