ICCF 2015-005: Eliminate Free Day Calculation
Michael Millstone, General Secretary
This proposal calls for the elimination of the Free Day provision in managing time controls in webserver games.
Proposal
This proposal calls for the elimination of the Free Day provision in managing time controls in webserver games.
Rationale
After completing an extensive player survey in 2014, some of the most pressing issues for players were the perceived problem with DMD and more accurate planning of start dates for tournaments to aid in managing games loads.
(a) With respect to the perceived DMD problems, subsets of players manage to complete entire games with 0 days remaining. As players are demanding that tournaments be completed faster and with less time controls, the practice of presenting and calculating time controls with a granularity of days has become obsolete. The webserver is capable of calculating and displaying time controls to the minute/second and reflection time should be a true representation of time remaining.
(b) As players are demanding accurate projections of tournament start dates, it becomes more critical to manage time controls accurately so that tournament end dates can be predicted with much better accuracy. Having the false day thwarts this ability to manage tournaments better, especially multi-stage tournaments, where one stage must be completed before the next can begin.
Other Comments/Considerations
None
Relevent Documentation
- ICCF Statutes – Article 3 (Time control is normally counted in days).
- ICCF Playing Rules Server – Paragraph 6 (Playing time calculation)
- ICCF Guidelines Webserver and Team Tournament Games – Section 6b (time control specified in days)
Voting Summary
- A vote of Yes will mean for tournaments starting on or after 01.01.2016; time controls will be calculated and displayed as an absolute number with days and minutes; remaining time will continue to countdown unless a player is on leave.
- A vote of No will mean no changes in respect to time controls and the slack in tournament planning and DMD conversations surrounding the use of the False Day calculation will have no merit.
- A vote of ABSTAIN is not a vote but means the vote holder has no opinion and does not wish to represent the correspondence chess players of his or her federation in this matter.
Comments by Relevant Committees, Commissions, or ICCF Officers
Eric Ruch (EB)
The free day dates back to the postal games, where time could only be counted in days, as the number of days between the day the player had receive the letter and the day the letter was posted. (It is clear that eliminating the free day will only apply to server games and that the postal games will always have a count in days).
Today, server enables us to know exactly time at which the move has been played, and there is no more "technical" reason to count in the reflexion time days. That will be a further step in the modernisation of the rules.
Eric Ruch (Rules)
Comment sent by Arjen Oudheusden (Delagate of the Dutch Federation) on behalf of Marc J.P.G. Schroeder ICCF ID: 370376:
"Fischer time control" is defined as follows. Before a player has made his move, a specified time increment is added to his clock. Time can be accumulated, so if the player moves within the delay period, his remaining time actually increases. For example, if the delay time is five seconds, and a player has four seconds left on his clock, as soon as his opponent moves, he receives the increment and has nine seconds to make a move. If he takes two seconds to move, on the start of his next move he has twelve seconds.
Currently FIDE sets a single time control for all major events: 90 minutes for the first 40 moves followed by 30 minutes for the rest of the game with an addition of 30 seconds per move starting from move one.
ICCF uses as standard time control: 10 moves in 50 days with duplication after 20 days with an addition of 24 hours per move starting from move one, the remaining hours are truncated to the day.
This is a variant of the Fischer delay, which would not truncate unused time to the day.
Mr. Ruch points out that the free day dates back to the postal games, where time could only be counted in days.
As shown above the free day can be defined non-historically as a (cut-off) Fischer delay.
The proposed elimination of the free day is motivated by:
(1) the perceived problem with DMD;
(2) more accurate prediction of tournament end dates.
Currently standard thinking time per 10 moves is limited not to 50 days, but to 60 days (when the free 24-hours are made explicit). The perceived problem with DMD is not the actual length of the game but the psychological attitude of the player who uses DMD. Several variants to shorten game duration have been investigated in the past. The accuracy of server time provides new possibilities. Spent time can be calculated in minutes (but of course displayed in a format like "days:hours:minutes"); remaining time will continue to countdown. In the spirit of FIDE one could formulate: 10 moves in 50 days with duplication after 20 days with an addition of 24 hours per move starting from move one. By the way, compared to the current situation this rule would lengthen the game by the non-truncated minutes. The gain would consist in simplicity as well as in the harmonisation with the FIDE way of thinking.
One could then reduce thinking time by imposing smaller numbers, like:
10 moves in 50 days with duplication after 20 days with an addition of 16 hours per move starting from move one,
or
10 moves in 46 days with duplication after 20 days with an addition of 16 hours per move starting from move one.
There are fundamental objections against the total elimination of any Fischer delay.
(1) Proposal ICCF 2015-005 perceives it as unfair that entire games can be completed with 0 days remaining. However, some players work during part of the day when they have no access to their computer. Take a player who is absent during 10 hours per day (8 work + 2 commuting). A randomly incoming move during his absence may waste him up to 10 hours of thinking time. As this may occur at each move before the next increment of 50 days, this player must keep a reserve of up to 100 hours. As long as a Fischer increment is added to his clock he cannot be driven out of his thinking time just because he has a job.
(2) In the case of postal games time can only be counted in days, as in general mail comes in once day, and leaves once a day.
In fact this feature also applies to server games. Moves cross time zones: a move dispatched in time zone A during daytime may land in time zone B during nighttime, when the receiver is sleeping. So in the case of server games the delays are not caused by the transmission of the moves, but by their reception. It is not reasonable to ask a player with - let's say - one hour left on his clock to stay posted at night near his computer to seize the last moment before he exceeds his time-limit.
(3) It is incorrect to state that acceptance of proposal ICCF 2015-005 would allow a more accurate prediction of tournament end dates. In multi-stage tournaments the end date of the stages which must be completed before the next can begin is fixed before the start of the tournament. As the end date is published in the start letter as well as on the internet, the prediction of tournament end dates is 100% exact.
A better question is whether a faster pace of the game would lead to fewer results by adjudication. As there is no complaint that currently the number of adjudications is unacceptably high, the proposal does not fix anything. In the case of DMD the game generally is lost to the DMD-player anyway.
Again: in the current system the prediction of tournament end dates is very accurate.
CONCLUSIONS
(1) The current rules should make explicit that a Fischer delay is applied. Stop defining the free 24 hours as an anomaly caused by history.
(2) Some time increment per move should be maintained.
(3) ICCF could consider standard Fischer delay (like FIDE), as opposed to cut-off Fischer delay (as ICCF does now).
(4) Playing time should be calculated in minutes (and displayed ad libitum).
Marc J.P.G. Schroeder
ICCF ID: 370376