Idea about Draws

[Brian_Mottershead]

As readers of this forum know by now, I have a problem with draws. They are part of chess, I know, but in the context of tournaments, I consider them a problem. They make it difficult to reliably determine a winner in a defined number of rounds, and they can make the outcome of the tournament dependent on agreements between players rather than properly contested games. It would be good for tournaments if there were no draws and all encounters between players were decisive, in my opinion. In a tournament, the purpose of an encounter between two players is to determine who is the better player on the day; so that the overall tournament can determine who is the best of all the players in the tournament, on the day. Draws work against these goals.

Most ideas for dealing with draws do not eliminate them entirely, or else they drastically change the game to the point where the players are no longer playing “chess”. Or else they rely on scoring gimmicks, so that two draws are worth less than one win, which seems unfair.

But here is an idea that recently occurred to me. I don’t suppose it is original with me, but I haven’t heard this idea before. Suppose two players arrive at a draw, through any of the usual methods (stalemate, repetition, 50-move rule, agreement, insufficient material, etc). The idea is this: when the draw is reached, the clocks are stopped, the initial position is set up again with colors reversed, and the clocks are restarted. (That is, the new game is started with as much time for each player as he or she had left on the clock from the previous game.) This is repeated until there is a decisive game, which might be by time forfeit (that is one of the players exhausts his alotted time for the encounter, and the other does not.)

There are probably details that would have to be worked out. For example what happens in a both-flags-down situation? But assuming those details can be worked out, with this system, all encounters between two players become decisive, though it might take more than one game, and more games might end by time forfeit.

Some questions that occur to me are: Would this system be allowed under the USCF rules? How would these encounters be submitted for rating – as one decisive encounter at the overall time control, or as several distinct encounters at different time controls?

[BobMessenger]

@Brian Mottershead

But here is an idea that recently occurred to me. I don’t suppose it is original with me, but I haven’t heard this idea before. Suppose two players arrive at a draw, through any of the usual methods (stalemate, repetition, 50-move rule, agreement, insufficient material, etc). The idea is this: when the draw is reached, the clocks are stopped, the initial position is set up again with colors reversed, and the clocks are restarted. (That is, the new game is started with as much time for each player as he or she had left on the clock from the previous game.) This is repeated until there is a decisive game, which might be by time forfeit (that is one of the players exhausts his alotted time for the encounter, and the other does not.)

Having the players play another game with reversed colors if the first game is a draw was used in tournaments in the 19th century, but the idea of carrying forward the clock times from game to game may be new.

@Brian Mottershead

There are probably details that would have to be worked out. For example what happens in a both-flags-down situation?

If you’re absolutely against the idea that a game can be drawn I suppose you could say that if both flags are down Black wins.

@Brian Mottershead

Some questions that occur to me are: Would this system be allowed under the USCF rules? How would these encounters be submitted for rating – as one decisive encounter at the overall time control, or as several distinct encounters at different time controls?

In my opinion the system would be allowed under USCF tournament rules but would be a major variation which would need to be announced in all pre-tournament publicity and announced and posted at the playing site. I think only the first game played in each series would be ratable, though, at the original time control and reported as a win, loss or draw, because all the subsequent games would be played at time odds. If the two players happened to start a game with the same time I suppose that game could be rated as a side game with its own time control, but that would be so infrequent that it’s not worth worrying about.

[BobMessenger]

@BobMessenger

@Brian Mottershead

There are probably details that would have to be worked out. For example what happens in a both-flags-down situation?

If you’re absolutely against the idea that a game can be drawn I suppose you could say that if both flags are down Black wins.

…
I think only the first game played in each series would be ratable, though, at the original time control and reported as a win, loss or draw, because all the subsequent games would be played at time odds.

If the first and only game in the series ended with both flags down, the result would be scored as a win for Black in the tournament standings but as a draw for rating purposes. One way to get the information you need would be to print two pairing sheets, in different colors, and maintain two “tournaments” in SwissSys: one for the tournament results and one for the rating results. You’d have to carefully explain this to the players and make sure that you didn’t mix things up yourself.

[Brian_Mottershead]

@BobMessenger

@BobMessenger

@Brian Mottershead

There are probably details that would have to be worked out. For example what happens in a both-flags-down situation?

If you’re absolutely against the idea that a game can be drawn I suppose you could say that if both flags are down Black wins.

…
I think only the first game played in each series would be ratable, though, at the original time control and reported as a win, loss or draw, because all the subsequent games would be played at time odds.

If the first and only game in the series ended with both flags down, the result would be scored as a win for Black in the tournament standings but as a draw for rating purposes. One way to get the information you need would be to print two pairing sheets, in different colors, and maintain two “tournaments” in SwissSys: one for the tournament results and one for the rating results. You’d have to carefully explain this to the players and make sure that you didn’t mix things up yourself.

If only the first game in the series is ratable, except in the unlikely case that one of the subsequent games gave both players exactly the same time, wouldn’t any kind of draw in the first game create a situation where you would have to keep two sets of books – one for rating purposes and one for tournament standings and pairing purposes?

[BobMessenger]

@Brian Mottershead

If only the first game in the series is ratable, except in the unlikely case that one of the subsequent games gave both players exactly the same time, wouldn’t any kind of draw in the first game create a situation where you would have to keep two sets of books – one for rating purposes and one for tournament standings and pairing purposes?

Yes, you’d need to keep two sets of “books”, as if you were running two simultaneous tournaments. Or you could simply say that the tournament was not USCF rated. It might cost you entries but it would be easier for you to direct and less confusing for the players.

[timjust]

Hmm…would this not effect the round schedule times?

[Brian_Mottershead]

@Tim Just

Hmm…would this not effect the round schedule times?

I think the main consequence is that you would have more encounters going down to the last few minutes of the round. This might lengthen the time between rounds somewhat.

[wzim]

I like this idea. But to make it work I think you would have to embrace it wholeheartedly. The major reason I like it is that it wouldn’t delay the rounds. There might be some time lost in resetting the boards. Pause the clock, turn the board around and start again. If it is Black’s turn to move Black goes first? I think this favors the stronger player because say that they were losing but managed to force a Draw, now materially they are back on even footing. Time is a factor because one player could be under 5 minutes and the other could have lots of time.
So no need to keep score etc. You wouldn’t look at it as seperate games but just the continuation of the Game.

On a side note Brian, wouldn’t something like this work with your scholastic idea of keeping the kids playing. Instead of pairing them again before all the games are in. You just have everyone play the whole round time.
If a player wins they keep the clock times and reverse the colors and keep playing until one of them runs completely out of time. Score table - Board 12 I won 3 and loss 2.

[timjust]

@Brian Mottershead

@Tim Just

Hmm…would this not effect the round schedule times?

I think the main consequence is that you would have more encounters going down to the last few minutes of the round. This might lengthen the time between rounds somewhat.

When would you suggest that the games that are drawn and have to be replayed with colors reversed be played (same time control?)? After all some of those draws could well happen many minutes or hours into the time control.

[BobMessenger]

@Tim Just

@Brian Mottershead

@Tim Just

Hmm…would this not effect the round schedule times?

I think the main consequence is that you would have more encounters going down to the last few minutes of the round. This might lengthen the time between rounds somewhat.

When would you suggest that the games that are drawn and have to be replayed with colors reversed be played (same time control?)? After all some of those draws could well happen many minutes or hours into the time control.

Tim, what he’s saying is that the clocks would not be reset, only the pieces. Say the initial time control is G/120. White uses 80 minutes and Black uses 90. In the second game the colors are reversed and White (who was Black in the first game) gets 30 minutes and Black gets 40. Repeat until someone loses or both flags are down.

[BobMessenger]

I’ve just thought of another situation where the game result would have to be reported differently for tournament purposes and rating purposes. If in the first game of a series a player’s flag fell and the opponent didn’t have mating material the result would be a loss for tournament purposes but a draw for rating purposes.

My recommendation would be for the tournament not to be USCF rated, at least until the format had been tried several times and players were used to it.

[wzim]

Brian what about at the Grand Master level where they in theory can draw at will? Wouldn’t a new tactic be to force the draw with Black so that they could have the White pieces? Then the opponent would force a Draw to get them back etc etc. So maybe with your idea of resetting, colors shouldn’t be changed.

[BobMessenger]

In my opinion colors should be reversed for each game and White should always move first. This gives White an incentive to play for a win because if the game is drawn he’ll have to play Black in the rematch.

[wilecoyote]

Note that the rating system requires both players to have the same amount of time for a game to be rated.

[BobMessenger]

@wilecoyote

Note that the rating system requires both players to have the same amount of time for a game to be rated.

Which is why I suggested that only the first game in a series could be rated (except in the unlikely event that the players happened to start with exactly the same time for one of the rematches).

[wzim]

@wilecoyote

Note that the rating system requires both players to have the same amount of time for a game to be rated.

Which is why I said the idea would have to be embraced wholeheartedly. One has to view the resets as a continuation of the game, not as seperate games. It is one game and both players have the same amount of time.
It is a fundamental change in that it removes the Draw from the equation totally. View the resets as overtime etc not as seperate games.

[BobMessenger]

@wzim

@wilecoyote

Note that the rating system requires both players to have the same amount of time for a game to be rated.

Which is why I said the idea would have to be embraced wholeheartedly. One has to view the resets as a continuation of the game, not as seperate games. It is one game and both players have the same amount of time.
It is a fundamental change in that it removes the Draw from the equation totally. View the resets as overtime etc not as seperate games.

If it’s a single game I don’t think it’s ratable, because it doesn’t follow USCF rules.

[wzim]

@BobMessenger

@wzim

@wilecoyote

Note that the rating system requires both players to have the same amount of time for a game to be rated.

Which is why I said the idea would have to be embraced wholeheartedly. One has to view the resets as a continuation of the game, not as seperate games. It is one game and both players have the same amount of time.
It is a fundamental change in that it removes the Draw from the equation totally. View the resets as overtime etc not as seperate games.

If it’s a single game I don’t think it’s ratable, because it doesn’t follow USCF rules.

Variation Announced in advance.

and for what it is worth I agree that it doesn’t follow USCF rules, it doesn’t even follow the rules of Chess as we currently play it. So at what point in the line does a Variation become unacceptable?

[BobMessenger]

@wzim

Variation Announced in advance.

and for what it is worth I agree that it doesn’t follow USCF rules, it doesn’t even follow the rules of Chess as we currently play it. So at what point in the line does a Variation become unacceptable?

I’m not sure exactly where the line is drawn, but I think this variation (multiple games reported as if they were a single game) would cross it. The same goes for any other variation where a draw is not a possible result, e.g. if it’s illegal to repeat a position for the third time.

Of course I’m just stating my opinion. I’m not the USCF Office, the Rules Committee, the TDCC etc.

[Brian_Mottershead]

@BobMessenger

@wzim

Variation Announced in advance.

and for what it is worth I agree that it doesn’t follow USCF rules, it doesn’t even follow the rules of Chess as we currently play it. So at what point in the line does a Variation become unacceptable?

I’m not sure exactly where the line is drawn, but I think this variation (multiple games reported as if they were a single game) would cross it. The same goes for any other variation where a draw is not a possible result, e.g. if it’s illegal to repeat a position for the third time.

Of course I’m just stating my opinion. I’m not the USCF Office, the Rules Committee, the TDCC etc.

It could be viewed as a “match” of multiple games, with each game at a shorter time control. Then the tournament is a series of these matches. However, because the players have different amounts of time in all but the first game, the second through Nth games in the match are at time odds.

This means, as Bob Messenger points out, that the second through Nth games are not ratable because odds games are not ratable. If for example, the tournament rule was to use an Armageddon game to break a draw, that would be similar. Armageddon is not ratable because it is a mix of time odds for White and draw odds for black, which is not standard chess, but the initial game is ratable. One also could not consider the encounter a single game, any more than game+Armageddon could be regarded as a single game.