Originally posted by CrowleyJust clarify this quickly, because I think I may be misunderstanding:
Just clarify this quickly, because I think I may be misunderstanding:
When you say "most points", "most wins" and "most moves in drawn games" - you are referring to the specific tournament, right?
What will happen in smaller groups like duals, 3 & 4 player groups? According to what I understand, then your points 1, 2 & 3 will be bypassed and both high ra p, or like Palynka proposed - the higher rated player should just be declared the winner?
When you say "most points", "most wins" and "most moves in drawn games" - you are referring to the specific tournament, right?
Yes.
What will happen in smaller groups like duals, 3 & 4 player groups? According to what I understand, then your points 1, 2 & 3 will be bypassed and both high rated players will just go through.
Only if they agree to a draw having played the same number of moves, which wont always be the case.
Shouldn't point 4 then be the virtual coin flip, or like Palynka proposed - the higher rated player should just be declared the winner?
I would prefer to see a virtual coin flip as the last tie break options rather than simply declaring the highest or lowest rated player the victor.
However, I wouldn't like to see a coin flip used in the final round of a tournament. But its use in earlier rounds would also help to discourage "arranged" outcomes.
Originally posted by GatecrasherThing is, these players are already 'bending the rules', so they will surely make sure that their move counts match up in order to bypass your solution.
Only if they agree to a draw having played the same number of moves, which wont always be the case.
I would prefer to see a virtual coin flip as the last tie break options rather than simply declaring the highest or lowest rated player the victor.
However, I wouldn't like to see a coin flip used in the final round of a tournament. But its use in earlier rounds would also help to discourage "arranged" outcomes.
This is why the 'coin flip' would be essential IMO.
The 'coin flip' would obviously only be used in a round, not a final.
Why should moves be a factor at all? If they're colluding to get two ties, then I'm sure they can also collude on the number of moves.
Personally, I don't like the randomness of the coin flip. I prefer a small, but non-random, bias against the "challenger" (i.e. the lower rated player). Like Phlabs said, this also helps (at least marginally) against sandbagging.
Originally posted by PalynkaI like Phlabs... though many of you DO NOT!
Why should moves be a factor at all? If they're colluding to get two ties, then I'm sure they can also collude on the number of moves.
Personally, I don't like the randomness of the coin flip. I prefer a small, but non-random, bias against the "challenger" (i.e. the lower rated player). Like Phlabs said, this also helps (at least marginally) against sandbagging.
P-
Originally posted by GatecrasherThis proposal does nothing to change the results in the scenario from the OP of this thread.
Only if they agree to a draw having played the same number of moves, which wont always be the case.
[Assuming, on games drawn after White's move, that both players get credit for the same number of moves played...] If the two top players in a group make draws against each other, then they necessarily have the same 'number of moves played in drawn games'.
Originally posted by SwissGambitOnly if the players in question don't draw any other games...
This proposal does nothing to change the results in the scenario from the OP of this thread.
[Assuming, on games drawn after White's move, that both players get credit for the same number of moves played...] If the two top players in a group make draws against each other, then they necessarily have the same 'number of moves played in drawn games'.
But I would also favour a coin toss as a final tie break measure in non-final rounds. So to restate:
1) The player with most points proceeds.
2) In case of a tie, the player with the most wins proceeds.
3) In case it is still tied, the player with the most moves made in drawn games proceeds.
4) In case it is still tied, a coin toss decides the player who proceeds.
So if you want to be sure of getting to the next round you have to do more than just draw against your main rival.
I have an idea, forgive me if its been said already, I don't have it in me to read 5 pages of bickering.
I suggest that the amount of points attainable from draws be limited.
1 point limit from draws in groups of 4.
2 point limit from draws in groups of 8
3 point limit from draws in groups of 16.
And so on and so forth. Players who are drawing games intentionally might be a bit more discouraged from doing so for they would be handicapping their potential point totals.
In the case of duals, allow players that tie to advance to the next round where they will play each other again. If they tie again, both are eliminated. Sure they might not like it but that's tough. Either don't tie intentionally or play better chess, no reason to hold up the show for the rest of us. I have an active dual tournament that was started in 09/07 and the second round still hasn't finished. Why? Because 8 of 32 groups tied and both players advanced. From the looks of things, at least 25% of the 2nd round will have tied groups as well.
I think both of these ideas are a bit more fair than a "coin toss" or favoring a higher rated player. Putting a point limit from draws in tournaments may do the trick. Players would be forced to save that point(s) for games they legitimately have to draw or end in stalemate.
As for my idea on duals, it may not be the best but it would certainly help expedite them.
Originally posted by GatecrasherWith a group with two players very much stronger [hundreds of points] than the rest of the group, it is highly likely they will both beat everyone else.
Only if the players in question don't draw any other games...
But I would also favour a coin toss as a final tie break measure in non-final rounds. So to restate:
1) The player with most points proceeds.
2) In case of a tie, the player with the most wins proceeds.
3) In case it is still tied, the player with the most moves made in drawn games ...[text shortened]... e sure of getting to the next round you have to do more than just draw against your main rival.
Under your proposal, a very strong player could get thrown out of the 2009 championship tournament after tying for first place in his group with a 15-1 score [again, it is possible that NONE of his games were drawn - he could have one win and one loss against the other very strong player] purely at random - and this the result of a policy designed to discourage quick draws.
Short, pre arranged draws cannot be stopped.
TD's since the famous (or infamous) Miles - Rubens no move draw
have not come up with an answer.
However, on RHP we do have something that TD's do not have.
A complete 'P.W.L. D.' record of games played on this site.
The player who has the fewest draws in their complete RHP record goes through.
This will most likely go against the player who been a subsriber
and supporter of this site the longest. Does that sound fair?
No matter what solution you come up with the sharp guys will find
a way around it.
Let it Be.
Originally posted by SwissGambitThat's a good point. One could let the random tie-break fall away in the case where that there are no draws... except two strong players could then collude to proceed by "winning" a short game against each other, rather than drawing.
Under your proposal, a very strong player could get thrown out of the 2009 championship tournament after tying for first place in his group with a 15-1 score [again, it is possible that NONE of his games were drawn - he could have one win and one loss against the other very strong player] purely at random - and this the result of a policy designed to discourage quick draws.
On the other hand, if one hasn't won a group outright, all the coin toss tie-break does is deny an automatic right to proceed.
This is not really a hard baked proposal, just some ideas to kick around.
Another layer of tie-break (between steps 2 and 3) could be inserted, rewarding quicker wins and longer losses (a bit like goal difference in soccer), effectively making the chance of a random tie break (except in the case of duels) quite remote, and the effect of short draws even more irrelevant. Moves in losing games minus moves in winning games, but with a cap on the number of moves counted in each game (say 40 or 50 to avoid abuse and players being rewarded for drawing out lost positions).
So of two players tied at 15 wins 1 loss in a round, the player most likely to proceed would be the one who won his games in the least number moves.
Originally posted by GatecrasherThis still encourages some game dragging...people will play out their lost games until the 'capped' move number, then resign.
That's a good point. One could let the random tie-break fall away in the case where that there are no draws... except two strong players could then collude to proceed by "winning" a short game against each other, rather than drawing.
On the other hand, if one hasn't won a group outright, all the coin toss tie-break does is deny an [b]automatic ri ...[text shortened]... player most likely to proceed would be the one who won his games in the least number moves.[/b]
However, it would also motivate players to look for forced mates and not just be content to ride their material advantage into an endgame.
It's better than a random determination.