ADM: Replace the current text of rule 28L2 with the following. (Note that the underlined text indicates the proposed change to rule 28L2 and is not intended as part of the formatting of the published rule.)
“28L2. Determination. In the first round, the bye is given to the player with the lowest USCF rating but not to an unrated player or a late entrant. In subsequent rounds, it is normally given to the lowest-rated player in the lowest score group but not to an unrated player. It is acceptable to give the bye to a higher rated player if doing so improves the overall color allocation for the lowest score group, subject to the limits specified in rules 29E5a and 29E5b. If there are no rated players eligible for the bye in the lowest score group, it is given to an unrated player who has played in a USCF-rated tournament too recently to obtain a published rating. If this, too, is impossible, a new player may be assigned the bye. New players should be indicated by NEW on the pairing card and wall chart. See also 28J, The first round; 28S Reentries; 28L4, Full-point byes after half-point byes; 29E5a, The 80-point rule; and 29E5b, The 200-point rule.”
My motivation for this ADM is the behavior of WinTD. I have had discussion with Mr. Doan (which may actually have taken place in this forum, though I’m not nearly patient enough with the search “functionality” to find it) in which we have disagreed on whether WinTD’s behavior is erroneous. Mr. Doan believes that selecting the player to receive a full point bye (paired out) is just a special case of selecting an odd player to drop to the next lower (nonexistent) score group. As such, it is allowed to select a player other than the lowest rated to treat as the “odd man” if doing so improves color allocation.
I can actually sympathize with this viewpoint, and this is in fact how FIDE Swiss pairing rules treat the bye. However, rule 28L2 is, in my opinion, an uncharacteristically clear and unambiguous rule in the Swiss pairing rules section of the Official Rules of Chess. As I read rule 28L2, I see no leeway for selecting a different player to receive the bye in order to improve color allocation. Nonetheless, one of the two major pairing programs used for USCF tournaments (and notably for all the national scholastic championships) does so.
(Please note that I apologize to Mr. Doan in advance if I have incorrectly represented his reasoning.)
I like this ADM. However, the wording “acceptable” makes it seem a TD can go either way and give the bye to the lowest rated player or to a higher rated player if that improves colors. Is this how you want the rule to be interpreted?
Also, it says “If there are no rated players eligible for the bye in the lowest score group, it is given to an unrated player who has played in a USCF-rated tournament too recently to obtain a published rating.”
It might be good to add a TD tip here that states the pairing programs can’t do this automatically since they won’t know which of the unrated players have played in a USCF-rated tournament too recently to obtain a published rating.
No, SwissSys would not have to be changed. I don’t think a variation is needed. I chose the wording “it is acceptable to give the bye to a higher rated player” to offer the option of doing so, not to mandate doing so.
What I’d really like is for WinTD to offer a configuration setting for strict adherence to the current 28L2 or to enable the current behavior, but in practical terms, that’s just not going to happen. It is far more likely that the author of SwissSys could be persuaded to put in an option to keep its current behavior of strict adherence to the current 28L2 or to allow the proposed behavior.
Since there are tournaments still being paired using pairing cards, the new rule should reflect that, at least in a TD tip.
In doing pairings where a bye must be given out, I have always gone with giving the bye to a rated player rather than to a new player or an unrated player so that they can get an official rating. This was a part of the old practice of doing pairings. Unrated players should get to play all of their games when they play in a tournament. This is especially true for one day, four round events. On the pairing card, I would write “No Byes” at the top of the card or put a star in a corner as a reminder.
I have seen organizers put “purchase” in the USCF ID for new memberships purchase on site. Since there are first-time players who will purchase a membership in advance, and since many TDs/organizers will take care of the new memberships during the tournament and add the IDs at that time, using the ID number to differentiate NEW is not a great option.
One other aspect I’ve seen of WinTD is that if only unrateds are in the bottom score group then it will often assign one of them the bye. One way around this is to assign the unrateds full point byes in rounds 11 and 12 of a five round event (since the length of the tournament isn’t entered in WinTD, it sees those upcoming byes and does not assign them the odd-player byes in earlier rounds).
PS If the tournament really is 11 or 12 rounds long then what I thought was the primary reason for not giving byes to an unrated (making sure they get their four rated games in so they can get an official rating) is most likely irrelevant by then.
We put markings next to the name (which also the players can see and often ask us about):
+m = needs USCF membership, already paid (if USCF ID is blank then new member, if filled in renew)
+m$22 = needs USCF membership, need to pay $22 (if USCF ID is blank then new member, if filled in renew)
if a player registers onsite we check membership immediately and mark the player if we need to process memberships. If they don’t have cash or check on-hand they also get the $, and told to come back with the $
Well, I’d still like to see the wording modernized a bit. Instead of suggesting “writing NEW on the pairing card”, why not suggest “specifying NEW in the pairing software or on the pairing card”?
It’s probably in a forum post—I can’t find a tech support e-mail thread about it. But yes, I looked at it as one of the cases where the rule book basically forgot to mention the caveats about the implications of the choice of “downfloat” for the remainder of the score group, both in terms of color and other “pairability” considerations. (If the bottom score group is at 1/2, it’s possible that some of the players have already met, and there are always team/club mate considerations).
While I can sympathize with this viewpoint, I note that the language of rule 28L2 (and even the numbering) has been the same since the fourth edition of the rulebook. In the third edition (which came into effect May 1, 1987), II.5.E states:
(Of course, II.5.e doesn’t say what to do if there is no rated player in the lowest-score group. Then again, the third edition rule book had a total of 196 pages, including an eight page TD checklist, five pages about the rating system, six pages of regulations for computer chess, 25 pages for the FIDE Laws of Chess, and 11 pages of other FIDE rules. Today, in the sixth edition, just the first chapter (which is what I think of as “rules for players” – how the pieces move, touch move, illegal move, drawn game, time controls – nothing about pairing rules) is 110 pages. Cue Carroll O’Connor and Jean Stapleton singing “Those Were The Days” with an out-of-tune piano …)
Unfortunately, I no longer have my copy of the second edition (dating from 1978), so I can’t check what the rule was back then. But, as May 1, 1987, was almost 28 years ago, I feel comfortable saying that it’s been the way it is for almost forever.
I’m really surprised there aren’t more instances of parents demanding to know why Johnny got the bye when he’s rated 103 and Timmy with a rating of 101 is playing …
Reading the rule literally means that if you have the following bottom score group (at 1/2 point)
A 750
B 700
C unr
where A and C have already played, then you would give the bye to B, and float both A and C up to 1.0. I think a simpler explanation is that the rule was written to really emphasize the importance of giving the unrateds as many games as possible rather than to describe in detail the process for pairing the bottom group.
Micah, you have just hit upon a very valid reason why I have pre-published for my scholastic events:
Current Ratings will be for pairing and section purposes at Chief TD sole discretion
and Terry, how much extra time do you require before the new pairings go up,
Secondly, I am curious here-do you still mail in tournament reports?
third, what about your wall charts, pairing charts, and up-to-date section results-- are these printed off the pairing
software?
I mention this because it is of EXTREME in building and maintaining a scholastic chess base that we get started on time, and get them out in a reasonable time period
If Terry turns around manual pairings as quickly as I used to then he could be manually doing everything spending no more than one more minute between rounds as he would using pairing software. The maintenance on the cards and wall charts are done while the round is still going on. The tentative pairings can be done before the final game is finished, with the pairing sheets already hand written and ready to go for all but a couple of score groups.
Computers make the final submission possible on-line. They handle the tie-breaks better if there are trophies (manual pairing often results in the first tie-break being cumulative).
I don’t mail in reports. I use the computer these days
I think the most I’ve ever paired by hand were under 100 players in different sections. But I think I was faster with cards than I am with the computer. Round 1, of course, was always late due to new registrations coming in after the deadline (you can’t turn them away - well I never could).
When doing them with cards, I was always working, and creating wall charts during the round, and all I’d have to do is fill in the results after the round was paired.
Subsequent rounds always started on time, or ahead of schedule if the players wanted. I never had any problems with pairing cards.
I haven’t used cards for quite awhile now, but I could. I’d pair by hand, do the paper wall charts, and result sheets, and then after the tournament, put the info in WINTD and send it on for rating. Sometimes I’d wait til Sunday to do that if I was whipped, but usually I’d get them out same night.
