Hi, Everyone. I’m about to run my first tournament, and I’m checking the rulebook for what to do in case of an unexpected withdrawal in a quad. I was just hoping to get some confirmation–or contradiction–from more experienced directors regarding my understanding of how to handle such a situation.
The hypotheticals below are based on chapter twelve of the rulebook as well as the pairing table on the quad tourney results chart. According to page 294, half is always rounded up, so half of a quad would be 2. Action is only required when someone withdraws before playing round two.
Player 1 withdraws before round two: No color reversals. Players 2 and 3 receive 1-point byes in subsequent rounds?
Player 2 withdraws before round two: In round two, player 4 receives a 1-point bye? In round three, 4 gets white instead of black; 3 gets black instead of white. Player 1 receives a 1-point bye?
Player 3 withdraws before round two: In round two, player 1 receives a 1-point bye? In round three, 2 gets white instead of black; 1 gets black instead of white. Player 4 receives a 1-point bye?
Player 4 withdraws before round two: No color reversals. Players 2 and 3 receive 1-point byes in subsequent rounds?
Any player withdraws after round two: No color reversals. The opponent of the withdrawn player receives a 1-point bye in the next (final) round?
I understand that withdrawals are to be avoided at all costs, but emergencies do arise. I can’t see any way of assigning points that doesn’t potentially punish at least one remaining player.
The bye-related sentences are all questions because I can’t find anything in the rulebook that deals specifically with byes in quads/round robins.
The thing about color reversals is just to use common sense, and figure out what the consequences would be for each player.
If player A is going to be deprived of a game because player B has withdrawn, simply look at player A’s assigned colors. If, due to the withdrawal, player A will end up with two whites and no blacks, or vice versa, then you should reverse player A’s colors in his next game.
If you use a system where all four players’ colors alternate in round 2, such as –
round 1:
1-4
2-3
round 2:
3-1
4-2
– then, if the withdrawal comes after 2 rounds, you have nothing to worry about.
Of course, if it is not possible to give any player his full complement of three games, that player should get a 1-point credit for the missed game. Whether you call it a “full-point bye” or a “win by forfeit” is not important.
You should struggle mightly to give each player all three of his due games (unless a player prefers to simply accept the free point). There are several ways to do this. You could jump into the tournament yourself (if your rating is more or less appropriate for the quad section you are jumping into), or there may be another player (USCF member) hanging around the tournament site but not playing, who could be pressed into service. Perhaps you could even get a player from a different quad to also jump into the wrecked quad, giving him an extra game or two. With any luck, there will be withdrawals in two quads, and you can do some quad-mixing to give everybody three games.
Usually a withdrawing player is one that has gone scoreless, so giving byes/forfeit-wins to the opponents is not an issue.
If somebody had to withdraw because of an unforeseen conflict (such as a husband whose wife has just gone into labor) then there is a chance that the other three opponents will have different scores against the withdrawing player and it could affect the prizes (example - the four played games in the quad drawn and two players getting forfeit wins/byes with the result that you have two 2-1, one 1.5-1.5, one 0.5-2.5). In such a case there will be some who say that the withdrawn player’s games should be removed entirely and put into an extra-rated-games section, so that the result of the quad/round robin would be determined only by the players who actually played a full schedule. The simplest answer is to just say that the player did play a rated game in the quad and the result counts - it may be unfair but it is unfair in a blindly unbiased manner (nobody knew in advance whether or not a player would need to withdraw nor which player it would be).
It should be pointed out that there is no requirement for a house player to be a USCF member, although if your house player has never been a USCF member you’ll be required to get an ID number for him. You will have to go through the membership exception process for the house player, but there will be no charge.
In a round robin, if a player withdraws before completing 50% of the schedule, the opponents’ results against the player are not counted for prize purposes. (However, the played games are reported for prize purposes.) So, if a player withdraws from a quad after the first round, you treat the event as a three player round robin, where each player sits out one round. The player who has already faced the withdrawn player has already had his “sit out” round. In this case, the maximum score will be 2.0.
(If a player has completed at least 50% of the schedule, his opponents’ results count, and the players who have not yet faced the withdrawn player receive a full point score for the unplayed game.)
First, congratulations (condolences? ) are in order. Welcome to TD Land!
Second, all of the advice given upthread is good. I will repeat pieces of it in this post.
In general, I think you should try to keep things as simple as possible. Basically, strongly encourage all your players to play every game. In most cases, especially for a small local event, that will usually suffice.
Another potential solution is to charge a small amount as a deposit, refundable after the start of the last round. Any player who doesn’t complete their schedule forfeits the deposit. I do this for larger round-robins that take multiple days…but for a one-day quad, this might not be optimal.
If someone does withdraw, try to press someone into service. You can, of course, be a house player in your own event if you choose - though I tend to not do that at my tournaments, because I know I’ll be distracted from running the event if I’m playing in it. Do NOT feel bad if you decline to play - that is entirely your prerogative.
If you have multiple quads with multiple withdrawals, you can at least make sure everyone gets a game.
First of all, thanks to all of you who have responded so far. I really appreciate the team effort.
Ken, I’m wondering where you found the rule that says to discard the game played with the withdrawing player for prize purposes. If it isn’t explicitly stated, maybe you could tell me which rule(s) allow you to make that judgement call.
Also, I assume you meant to say, “…the played games are reported for rating purposes,” not “for prize purposes.”
thank you! With all the work I’ve been doing, I must have forgotten all about 30B. that definitely takes care of at least part of the issue. The other part is most likely taken care of using 22B–Full-Point Byes, from the section of the rulebook not dealing specifically with Swiss pairings.
Phew. I think I got it, but check my thoughts below.
Withdrawal before round 2: For prize purposes, pretend the guy never showed up in the first place (but rate the game and use crenshaw-berger tables to deal with colors in last round). According to 22B, each of his next opponents receives a full-point bye, although the final scores would be practically the same without awarding the point. Even if withdrawal guy drew or won his first game, that result is not counted, even for his opponent, for prize purposes.
For the withdrawn player’s first-round opponent, I pretend he also had a full-point bye. Fill out results sheet as normal. I attach a note of some kind that reminds me to nullify the withdrawn player’s opponent’s round 1 score and award a full-point bye for prize purposes only.
Withdrawal after round 2, give a full-point bye to the guy’s opponent in round 3. Award prizes as usual. Guy who withdrew is not eligible for prizes (32C1), but his games are still part of his opponents’ records for prize purposes.
This is incorrect. Do not give “full point byes,” as they do not apply in this situation. You are running a three-player round robin (and there has been one “side game” that will be reported separately). There is no score for each player’s “sit out” round. The maximum score in a three-player single round robin is 2-0. (Note: What would happen if the first opponent did not win the game? If you give “full point byes” to the remaining opponents, you place the first opponent at a disadvantage!)
Technically, you’re not giving a bye. Instead, the game is an unrated forfeit win – “X” for the opponent who is present and “F” for the player who has withdrawn. But giving a “B” (full point bye) to the opponent and a “U” (unplayed round) to the withdrawn player will have the same effect. The maximum score in this case is 3-0.
This appears to be an easy source of confusion. In a round robin, there really aren’t full point byes. If there is an odd number of players (or the withdrawn player completed less than 50% of the schedule), the “sit out” round is not a bye; everyone has one of these, so it’s nothing other than a break in the schedule, a chance to go get coffee, or a meal, or rest for the next round. If the withdrawn player completed more than 50%, the unplayed games are treated as “no-show forfeits.”
I promise I’m not trying to be difficult, and I appreciate your help! I’m just totally new to this and don’t have lots of directors available in the area to help me gain experience.
Okay. So with a withdrawal before round 2, I don’t quite understand how it follows that all of a sudden I’m running a 3-player RR. I understand from a practical point of view, but then why do the Crenshaw-Berger tables exist? In the case of a withdrawal before 50% of games have been played, I’m supposed to use the tables to alter color allocation. If I just change the event altogether and make it a 3-player RR, the pairing numbers could change, and then what am I supposed to do for colors? It also raises questions about how to finish filling out my QUAD results sheet…
Why can’t I just give the opponent of the withdrawn player a bye for prize purposes, ignoring his round 1 result? Then I keep the original color allocations in order for reversals, if necessary. Everyone receives a 1-point bye under 22B, which is not specific to the Swiss system.
30B states, “Scoring is the usual…except that players who withdraw before playing half their scheduled games shall be scored as not having competed at all. Their completed games must still be rated, but they are not considered part of their opponents’ records for prize purposes.” From the italicized segment, I deduce, for prize purposes, that there were an odd number of players in round 1. Hence the 1-point bye under 22B.
I submit my original quad results sheet with the real results from the round 1 games and include only 2 byes (not all 3, as one was only for prize purposes) for the players who would have played the withdrawn player.
OK, I think I see what’s missing here. Thought experiment: Suppose you had to assemble the pairings for a seven player round robin. How would you do that? The answer is that you would use the eight-player table and treat player number 8 as a “ghost.” So, you would assign pairing numbers 1 through 7 randomly to the seven players. Whenever a player is paired against player 8, that is the player’s “sit out” round. Then, you would fill out a round-robin pairing table for seven players exactly as you would for eight, but there is no eighth row or column. Also, note that by treating player 8 as a “ghost,” all players get exactly three whites and three blacks.
Now, what’s the difference between this thought exercise and the case where you are running a quad in which one player drops out? Easy – the player who drops out becomes the “ghost.” But, you have to be a little careful about color allocation. What really happens is that, instead of tossing for color in the last round, you assign the colors so that each player has one white and one black. (Here, you are ignoring the game that the “ghost” played against one opponent for color purposes. That game has now turned into a “side game,” which you will enter in a separate section for the rating report. Just name that section something like “side games.” Note that there is no problem at all if one player appears in more than one section of a rating report.
According to the section of rule 30 titled “Quads” (sorry, I don’t have my rule book in front of me, so I’m doing this from memory – it’s a favorite party trick ), the pairings for a quad are:
Round 1: 1-4, 2-3
Round 2: 3-1, 4-2
Round 3: 1-2, 3-4 (toss for colors in this round)
(Note that, unlike larger round robins, you assign the pairing numbers 1-4 in rating order, not randomly.)
Now, here is how you handle color allocation in the third round, depending on which player withdraws after the first round.
If 1 withdraws: 3 has white, 4 has black, and 2 sits out. (Then 2’s color history is WBx, 3’s is BxW, and 4’s is xWB, where “x” indicates the unplayed or ignored round against the “ghost.”)
If 2 withdraws: 4 has white, 3 has black, and 1 sits out. Then 1’s color history is WBx, 3’s is xWB, and 4’s is BxW.
If 3 withdraws: 2 has white, 1 has black, and 4 sits out. Then 1’s color history is WxB, 2’s is xBW, and 4’s is BWx.
If 4 withdraws: 1 has white, 2 has black, and 3 sits out. Then 1’s color history is xBW, 2’s is WxB, and 3’s is BWx.
Yes, everyone’s final score will be inflated by one point, but everyone will be inflated equally. You will still have to handle colors exactly as in the above case. The only difference is that you are adding one full point to everyone’s score. You can go ahead and do that if you want. The effect is the same.
No! This is exactly where the train derails! I think you are insisting that you must assign a score to each player for each round. This is not true of a round robin. It ain’t no Swiss. For the “sit out” round, the player just sits on his hands for that round, and absolutely nothing happens to his score – zilch, nada, nichts. The point is that each player is treated equally in that there is one round that is a “non-entity” for their tournament performance. The only difference is which round is the “sit out” round for each player. What that also means is that, while the tournament is in progress, different players have completed a different number of games (but the difference is just one, depending on whether the player has had his “sit out” round or not).
It probably helps if you’ve seen a round-robin crosstable before. Here is a round robin cross table for a fictitious five-player round robin:
1 2 3 4 5
1 Attack, Arthur x 1 1 1/2 0 2 1/2
2 Bishop, Ben 0 x 1/2 1/2 1 2
3 Castle, Cathy 0 1/2 x 1 0 1 1/2
4 Diagonal, Dan 1/2 1/2 0 x 1/2 1 1/2
5 EnPassant, Ellen 1 0 1 1/2 x 2 1/2
Ah, I think this is another source of “hang up.” You’re trying to submit the rating report with just a single section. Sure, you can do that, but then your cross table will not reflect the actual results if the opponent of the withdrawn player did not win the game. (You will have given a full point for free to the other two players who did not face the withdrawn player, but the player who did face the withdrawn player will have his actual score, which might be a draw or a loss.) I very strongly encourage you to submit that one game in a separate section as described above. The rating fee will not change, as it is based on the total number of games in the event. (I assume that if you have several quads, you are also planning to submit each quad as a section of a single tournament, if for no other reason than you would be paying $3.00/quad in rating fees instead of $1.50 if you submit them separately.)
By the way, you write that you will “submit your original quad results sheet.” Do you actually plan to submit the rating report on paper to the USCF by mail? I would very strongly encourage you to submit your rating report through the TD/Affiliate web site. You can manually create a rating report and add four player round robin sections, one at a time, to the report. You’ll save a good chunk on the rating fee, and your event will be rated much more quickly. That tends to lead to player (“customer”) satisfaction.
Okay I might actually have it now. I’ve spent awhile digesting everything you wrote and practicing with invented scenarios. The part that was confusing me so much was the pairing numbers. I didn’t see how, if we were going to just start an entirely new, 3-person crosstable, we could make the numbers work. It seems that we just pretend that the quad is still going on for pairing purposes, so everyone keeps the same pairing numbers, even if that is not what it looks like on the new crosstable.
First thing to do is create a 3x3 crosstable and transfer the game scores from round 1 of the players who both stayed into round 2. The game played with the dropout can go on another sheet of paper all by itself later on.
Then pair round 2 just as if everyone still had the same crosstable setup and use the same pairing chart. Whoever is paired with the absent player simply does not play. Nothing is recorded for that player in round 2 as there is no space on the crosstable anyway.
In round 3, continue to follow the pairings as if the quad were still in effect, and use your brain–or the crenshaw-berger tables–to determine whether reversals are necessary. Whoever is paired with the absent player sits out. Nothing is recorded for that player in round 3.
What’s the big deal? If you give full point byes to everyone, everything balances out.
If the withdrawing player lost his one game, there’s no need for a separate “side games” crosstable. Just include that result in the regular 4-player crosstable. Give the other two players either full-point byes or forfeit wins. Everything comes out the same. Give the withdrawing player either two forfeit losses or two unplayed games (“U”). It makes no difference.
If the withdrawing player won or drew his one game, then, in order to keep the “results crosstable” (used to award prizes) in sync with the “ratings crosstable” (the one submitted to USCF), you could then, if you prefer, use the “side games” technique. But there is really no need. You could still include the odd result in the 4-player crosstable, and submit it that way to USCF, while the version posted in the tournament hall could show just 3 players, 2 rounds.
In any case, however, colors need to be handled differently than if there had been no withdrawal. But this is not complicated. In round 2, pair as normal (equalize colors). In round 3, do not toss for color – instead, equalize colors for the player who has played only one game. As it turns out, this will also equalize colors in the two “real” games of the player who will end up with three games:
If player 1 withdraws:
round 1: 1-4, 2-3
round 2: 4-2
round 3: 3-4 (don’t toss – equalize colors for player 3)
If player 2 withdraws:
round 1: 1-4, 2-3
round 2: 3-1
round 3: 4-3 (don’t toss – equalize colors for player 4)
If player 3 withdraws:
round 1: 1-4, 2-3
round 2: 4-2
round 3: 2-1 (don’t toss – equalize colors for player 1)
If player 4 withdraws:
round 1: 1-4, 2-3
round 2: 3-1
round 3: 1-2 (don’t toss – equalize colors for player 2)
Of course, you’re exactly right, provided you remember to also give the player who actually played a game against the withdrawn player a full point bye for that game. That’s what I was trying to explain when I said that “you would then inflate everyone’s score equally.” Of course there’s nothing wrong with this, but it then means you would effectively have a three-player round robin with scores between 1.0 and 3.0. It just looks strange.
Mostly, I was trying to explain that, in a round robin with an odd number of players, not having an opponent for a round does not mean the player has to be given a full-point bye. That’s what happens in a Swiss, but round robins are different, and most directors just don’t have experience with round robins.
Yes, because in that case the opponent of the withdrawn player scored a full point, which is the same score he would have had from a full point bye.
Well, it makes no difference from the rating system’s point of view, but it can make a difference in the MSA crosstable. For instance, suppose player 4 withdraws after the first round. In the first round, 4’s opponent is 1. Suppose 1 loses that game, 1 draws against 2, and 1 beats 3. Also suppose that 2 and 3 draw. Then, if you leave player 4 in the crosstable and assign full point byes for the unplayed games, you will have the following:
1 2 3 4
1 x 1/2 1 0 1.5
2 1/2 x 1/2 B 2.0
3 0 1/2 x B 1.5
4 1 U U x 1.0
Who won the tournament? Someone looking at the crosstable in MSA would conclude that player 2 won the tournament, and players 1 and 3 tied for second. In fact, player 1 won (1.5/2), player 2 was second (1/2), and player 3 was last (0.5/2). As you say below, it doesn’t matter, since the unplayed games are not rated. However, some people make a big fuss about MSA crosstables, thinking of them as a historical repository.
This is perhaps a stretch of a hypothetical, but what if this is a scholastic quad, and the prize is a first place trophy? Do you want to be the TD who has to deal with player 2’s parents complaining that their child did not win the trophy even though MSA clearly shows their child had the most points? For me, the answer is “no, thanks.” At the risk of appearing stubborn, I strongly prefer placing the one game in a “side games” section, even if the withdrawn player lost that game.
Yeh, I guess if you’re worried about historical use (or mis-use) of MSA crosstables, you’d want to have a separate “side game” listing if the withdrawing player won or drew. 9 times out of 10, though, he will have lost, and you can avoid the mess.
Even with multiple-section Swisses, I don’t like seeing “side games” cluttering up history. If there is a cross-section pairing, usually the game can be included in the crosstable of the winning player’s section. If there is a house player, the game can normally be added to the “real” player’s crosstable.
Also, if anyone joins the USCF at the tournament, you should have them pay in cash, by credit card, or by a check made out to you (not the USCF). You will be able to enter their credit card payments through the TD/Affiliate website, or if they’ve payed you in cash or by check, you will be able to put the charges on your own credit card, but if they make out a check to the USCF, processing of their membership and (as a consequence) the tournament results will be delayed while their check is being sent through the mail to the USCF.
Held the quads yesterday! In the end, we had an even 12, so a great start. No withdrawals and no huge problems. Most people had a good time and thought it was well organized. I’m really glad to be able to provide rated chess opportunities for people around here. Thanks, Everyone.
) are in order. Welcome to TD Land!
), the pairings for a quad are: