National Scholastics: Repairing Round One

The issue of whether to repair players in the National Scholastics has sparked a little debate recently. Repairing players is done to give the maximum number of players the chance to face real opponents, especially at tournaments like the Nationals where players and teams may have traveled far and wide for the opportunity for rated competition against a uniquely diverse group of players. An additional minor benefit for the tournament is that the potential number of perfect scores at the end is also reduced with every repairing of two players who would otherwise receive a forfeit win.

The downside to getting a forfeit win is that the player gets no tiebreaks for that round. If he or she was in the top half of the tournament, he would probably have won the game against a real opponent anyway, and then would receive tiebreak points for that game as well.

Maybe a slight adjustment in the tiebreak rules might help. In the first round (the only round to allow repairings), a forfeit win receives the same tiebreak points as a player scoring an even score. In the Nationals, that means that players who get 1-F in Round 1 would get 3.5 Median tiebreak points, the same tiebreaks as if their first round opponents scored 3.5-3.5. In the normal first-round pairings, that might not be that far off from the averagae tiebreaks for the opponents of the players in the top half anyway.

Under this system, the players in the bottom half might be expected to get more tiebreaks from their higher-rated opponents than just an even score. but the lower-rated players who win on forfeit in the first round would not have been expected to win against a higher-rated opponent.

This would need to be coded into the pairing software, but that shouldn’t be insurmountable.

This doesn’t address the issue of whether repairings should be done in the first place (hopefully this doesn’t mean that the first post to a topic is off-topic), but at least it offers a slightly bigger upside if you wanted to consider not doing them.

  1. I’m curious as to what the objection to repairing is. Obviously you want as many games as possible decided over the board. Is there some strong counterargument?

  2. “This would need to be coded into the pairing software, but that shouldn’t be insurmountable.” I assume you are not proposing this for all tournaments, just the special-case National Scholastics. You want the software makers to add an additional tiebreak system to the programs just for that? (“Modified Modified Median”?) Talk about wagging the dog.

I’m not a fan of a system which would give tie-break points to someone who wins on forfeit. One purpose of tie-breaks is that when it comes to trophies, to reward the player who played the tougher schedule. Other than having to wait around doing nothing, it doesn’t get any easier. There’s no risk of losing which the higher rated player otherwise has.

Giving credit for beating a player who scored 50% to someone who won on forfeit while giving someone else who beat someone over the board less because he played someone in the bottom half who played like a bottom half player favors the wrong player on tie-breaks.

If two top half players should face each other in the first round due to repairing, each player will receive better tie-breaks from that opponent if the top half players play like they belong in the top half for the remainder of the tournament.

Looking ahead to following rounds, the player that lost the first round repaired top-half vs. top-half game, receives an easier second round pairing. It would be comparable to switching the order of the first 2 opponents.

Steve’s proposed modification creates more problems than it solves.

But the same principle is already used for players whose opponents drop out of the tournament later.
Under Median tiebreaks, a player in a seven-round tournament who loses in round 1 and whose first-round opponent drops out of the tournament in Round 2 would receive 4 Median tiebreak points for that opponent (1 point for the opponent’s first round win plus one half point adjusted score for each of the six remaining unplayed rounds). Conversely, if the player had instead won his game in Round 1 against an opponent who then dropped out, he would receive 3 tiebreak points for that opponent (0 points for the opponent’s first-round loss plus the same six half-points adjusted score). If there were no such adjustment, then players whose opponents drop out of the tournament would really be penalized by receiving no tiebreak credit for having played them.

Of course, a forfeit win is even more of a gift than if your opponent allows Fools Mate. But if the player won his first-round game by forfeit against a no-show opponent, instead he now receives 0 tiebreak points for that player. That seems to be somewhat inconsistent with the idea of not being penalized on tiebreaks by an opponent who drops out of the tournament.

Steve_Immitt wrote:

If I understand Steve’s argument correctly, then the appropriate tie-break points against a no-show in a 7 round event would be 3, not 3.5. To award 3.5 for having 1F against a no-show implies that somewhere in the next six rounds, that same no-show would be given credit for a win to bring him back up to 50%.

While I see some logic in Steve’s argument, to me it still comes back to the fact that a 1F-0F has no risk attached. Why should someone who has no chance of losing get more tie-break points than someone who had to take that risk to obtain victory?

The vast majority of round 1 games are blowouts. Unless they use accelerated pairings, typically you see A players against E players. There isn’t that much risk involved in playing the game, unless your kid didn’t sleep well or doesn’t respect his opponent (admittedly a problem in scholastics). It may take longer to wait for a forfeit than to actually win a game.

Michael Aigner

The objection would be from higher rated players who get repaired against (much) higher-rated opponents in the first round, while their rivals are paired down against lower-rated opponents. But it’s not so simple, because for every player who complains about being repaired against a higher-rated opponent, there could also be yet another higher-rated player who would complain that he was not repaired, and is thus being depirved of tiebreak points by getting a forfeit win, even though the TD knew that his opponent was not coming (an opponent who was mis-paired in the wrong section, and then was removed after the pairings were posted, or who had previously requested a bye for that round, for example). But if a forfeit win yielded some modicum of tiebreak points, the latter complaint would probably not be as common as the former.

In his first article on this tournament Mike Klein wrote about the pairing problems, ending with, “Stay tuned - the novella’s ending may not have been written yet.”
In his second article this is all he had to say, “The round one pairing fiasco, detailed in the previous report, mercifully ended.”
I am reminded of the Saturday morning action programs I watched as a child. At the end of one program, the hero would be knocked over a cliff. One would have to wait a week for this ‘cliff-hanger’. The next Saturday the program would begin with the Lone Ranger landing on his horse and riding off…
Michael Bacon

I can see why some might complain, but they really don’t have a leg to stand on. No one has a “right” to a forfeit win. As for your proposal, I have a feeling you’re attacking the wrong problem. Do we want to make forfeit wins more attractive? If you want to tinker with the tiebreaks to deal with this, why not simply drop the first round?

  1. I’m curious as to what the objection to repairing is. Obviously you want as many games as possible decided over the board. Is there some strong counterargument?

  2. “This would need to be coded into the pairing software, but that shouldn’t be insurmountable.” I assume you are not proposing this for all tournaments, just the special-case National Scholastics. You want the software makers to add an additional tiebreak system to the programs just for that? (“Modified Modified Median”?) Talk about wagging the dog.
    John Hillery
    westernchess.com

John, From the blogs I have seen on this general subject, the thought a few coaches have is that
it is preferable from a team perspective to get a point for a no show in the first round rather than
have a player, particuraly the lower rated player in the matchup, take the chance of loosing.
The players development is seen as truly secondary to the overal good of the team, which is
indeed unfortunate.

Rob Jones

I hate it when I get a forfeit because I almost automatically lose any tiebreakers for position I might have had. Should I really be punished because my opponent didn’t show up, something I had no control over? Besides I came to the tournament to play chess, not get a no show win. Repairing sounds fair to me, but it has never happened in the tournaments I’ve gone to.

by Windrider on Wed May 19, 2010 6:38 am #190882
I hate it when I get a forfeit because I almost automatically lose any tiebreakers for position I might have had. Should I really be punished because my opponent didn’t show up, something I had no control over? Besides I came to the tournament to play chess, not get a no show win. Repairing sounds fair to me, but it has never happened in the tournaments I’ve gone to.

Thats the spirit!!!

Rob Jones

One interesting thing to note is that Daniel Gurevich (2136) and James Black (2070) tied for first (at 10.5-1.5) in the K-6 blitz at the National Elementary.

Daniel Gurevich played on board one throughout and James Black took a tie-break hit by having a round one forfeit. Since the first tie-break was modified median James’ forfeit win was thrown out along with Daniel’s win over a 6-6 performing player, leaving James as the winner (in spite of the forfeit win effect on his tie-breaks) with 45.5 tie-breaks versus Daniel’s 40.5 ( in the final round James’ four opponents from the earlier rounds went 7.5-0.5, all counting in Modified Median, while Daniel’s six opponents went 2.5-9.5, with only 1.5-8.5 counting in Modified Median). That six point pickup gave James his come-from-behind tie-break margin on modified median without having to worry about the Solkoff tie-break (where Daniel still had a 46.5 vs 45.5 edge because James’ forfeit win really hurt him there).

What would happen if the first round repairing deadline had passed, and the only players who had not arrived were Player #1’s opponent on Board 1, and Player # 2’s opponent on Board 2?

Try to find a couple of house men to plug in. But assuming those two did not want to play each other and the TD went along, why should they be rewarded for this with a tiebreak bonus, as your proposal would do?

It really shouldn’t matter whether you are talking about #1 and #2 or #400 and #401. The policy should be clear and apply to everyone equally.