1. The technical rules of each 1v1 should be uniquely established by agreement of the two players involved.
"technical" means any details not explicitly mentioned in these guidelines
2. Each match should consist of X* number of rounds. Each round is a battle that starts with both players on full health and ends with one player dead. Players may not change equipment or any game settings while conducting a match.
*X should be twice the value of an odd number greater than 1 (eg. 6, 10, 14). The starting conditions of each round should be identical (same map, same obstacles, etc.)
3. Every match shall be proctored by an judge agreed upon by both players. The judge has the right to interrupt a round if rule violation is suspected*. The judge can also invalidate a round that lasts for longer than 5 minutes
*If the judge accuses a player of cheating, the round is restarted and the accused player is allowed to convince the judge of his/her innocence in a reasonable way. If the judge is unconvinced, the opposing player can choose to waive the violation. By waiving the violation, the opposing player also waives any rights to outlined in Rule #5. If there is no waiver and the judge is unconvinced, the violator forfeits the round.
4. Half of the rounds in each match shall be done on the EU server, and half on the US server. Winning the majority* of the rounds in each server earns half a point. This means the score of each match could be 1-0, 0-1, or 0.5-0.5.
*If X is chosen appropriately, there should always be a majority winner in each server.
5. The loser (and strictly the loser) of a match can accuse the winner of violating the rules. To make an accusation, the he/she must say specifically what was done against the rules. If the judge deems the accusation reasonable, the winner must offer video proof or a concessional rematch* in return.
*In a "concessional rematch" the loser is allowed to engage in the actions specifically described in the accusations. The winner is expected to follow the old rules. If the loser achieves a better score in the concessional rematch, 1 point is added to his/her old score. If the loser fails to do better in the rematch, 1 point is subtracted.
6. Several* matches must be conducted. In each match, both players must have identical loadouts and game settings. Different loadouts are used for different matches. The loadouts used must be established as prior agreement between the two players.
*Using a larger variety of loadouts gives a more accurate comparison of the players' holistic skill.
7. The scores from different matches should be averaged* to yield each player's total score. The player with the higher score is deemed holistically better than the other
*Averaging puts more value in consistent performance over all situations rather than high performance in select situations.
"Fair," does not exist. To me, it's way too idealistic. Though the meaning of fair, like most things in a communicative language such as English is subjective ( which is why I dislike it and it also happens to be my worst subject at school, at the moment.. ).
Anyway, when conducting experiments, you need controlled variables as well as independent variables.
The independent variables are the people ( also known as play styles ) participating in the challenge.
The controlled variables are needed so we have fair results.
Controlled variables are :
Latency - They must be exactly the same at all times. Server does not matter as long as both player's latency are the same.
Equipment - The same. UV, Status, Type, Damage, Range, Etc.
Frame Rates - Must be the same at all times.
Keyboard setup - Some are easier to use than others, they must be the same.
AT/AA - Both must have it either on or off. As long as they both agree on the same, it should be fine.
Monitor - Should be the same, long hours from multiple matches may put stress on eyes.
Mouse - Should be the same DPI, all buttons should work. Both mice should be the same.
Multiple matches - Eliminates outliers. The more matches, the better, the more matches, the closer you get to real results. Then it should, as you said, be averaged.
Fair sounds like perfect. Perfection doesn't exist, but this is probably the closest you can get to it - the closest you can get to having a, "fair match."
Just let people use and do what they like, it's all fair on a technical view from the server's side of things - the game's rules, unaffected by our own nature and morals since we all know how perfect those are.