Players online, especially juniors, can be tempted to break the rules by getting assistance from a computer engine. We need to put measures in place to protect the vulnerable members of our community.
Our Fair Play technology is implemented to protect players with exceptional performances from accusations of unfair play.
And, to support vulnerable players who may not have sufficiently well developed self-control mechanisms with early reminders to make better choices in their daily lives and when playing chess on Tornelo.
Analysis of games by a leading chess engine demonstrates a higher move-correlation indicates the more likely it is a person is using engine assistance. The last thing we want is for players to be banned from participating in a sport they clearly enjoy and love. Our goal is to achieve fair play for everybody.
We require all players to use their real names and verified genuine identities in order to play on Tornelo. Anyone found to be playing under a pseudonym or false name will be removed from the site. Identity verification is done by the Organizer of your event; Tornelo will never request any identification documents.
Unfair Play is Prohibited
We define this as using any external assistance to gain an unfair advantage over your opponent. Some examples might include computer engine assistance, opening books, endgame tablebases or asking another player for help. This is not an exhaustive list of what we deem Unfair.
Our expectation is that all players in an Organizer’s community are behaving ethically and treating their opponents fairly at all times.
We always start from a Presumption of Innocence.
Despite our expectations and assumpions, we must have a process in place to identify and penalize players who undermine the enjoyment of community members by obtaining an unfair advantage in some manner. This is our process:
We encourage Organizers to investigate players using the following critera:
These are temporary measures. We are currently putting technology in place to analyse the games of all players in every event and providing the Arbiter with real-time reports on game correlation analysis.
Once a player has been selected for observation we will complete an analysis of all games they played in that event.
We submit all games for analysis to Dr Ken Regan, a renowned statistician and world leading expert at detecting assistance in chess games. This first Screening Test asks the question: "Is this player performing outside “normal” expectations?"
The analysis provides an Outlier Index where 50 is expected for the rating, 51-60 is normal, 60-70 is of concern and 70+ is well outside "normal" expectations.
If Dr Regan is unavailable, we use our own analysis technology to ask the same question: "Is this player performing outside “normal” expectations?"
Our methodology looks at two measures;
- Centi-Pawn Loss (CPL) as a measurement of performance compared to that of a computer engine. A lower CPL means a game more closely matched to that of a computer.
- MoveMatch % (MM%) as a measurement of how often a move matches the first choice of a computer engine. A higher MM% means a game which is more closely matched to that of a computer.
Our decision criteria (for players at 1600 rating) shows you are playing OUTSIDE your normal range with results of either:
CPL Score below 20.
MM% above 60%
These benchmarks are adjusted in line with individual player ratings.
As soon as games are identified as being Outside the normal range of that player ability, Event Organizers will notify:
These parties will be notified that an investigation is underway and that one (or several) games were flagged as being at a level which is likely to indicate they were played with Computer Assistance.
We realise that people (young players in particular) may occasionally make poor decisions. All players are provided with an Amnesty Period where they will be given an opportunity to either:
If a first-time offender admits to having received assistance:
If a player asserts their games were played without assistance, all Likely games will be submitted to Dr Ken Regan's DEEP Analysis. If this is not available games will be manually reviewed as follows:
We may, at this stage, review ADDITIONAL games that this player has played in order to create a more reliable expectation of performance.
If the panel of Arbiters/Coaches determines the player has, without a doubt, received assistance:
These sanctions are subject to appeal, see 7. below.
Where this process relates to someone Under the age of 18, all details of proceedings will be shared with Parents and/or School Teachers. Otherwise details will remain confidential between:
Event arbiters will be required to maintain confidentiality of this information at all times.
There must be checks and balances in all sanctioning processes to minimise the number of "false positives".
A player may appeal, within 7 days of the decision, by:
Tornelo's Fair Play Officer will require the Organizer to submit evidence that has led to the decision and will also collect independent evidence.
The Bond will be returned in the case that the Organziers' decision is overturned, but will be forfieted if the decision is upheld.