Batumi Olympiad

[Chris Rice]

Good suggestion for a new tie break system. 'One potato, two potato'.

[Richard Bates]

Ah, I get what you mean now. I'll leave you to do a relative comparison over the last 30+ years.

Data from Olimpbase.

In Olympiads at least, the period from 2004 to 2014 was dire in terms of finishing positions.

Which begs the question of what changed, what is going right and what can be done to continue this

And then, to bring back to your other point, what lessons can be learned from and for the Women's team

Not compromising your chances by resting your best player* at a critical point would be a good start.

*accepting that there may be extenuating circumstances of which we are not aware.

I unilaterally withdraw my comments on this issue and would like to apologise to all concerned.

[JustinHorton]

The Chinese players diverge on move 17

Shankland was interviewed after the game and it turns out the the USA team had prepared this opening together. In the Wesley So game Shankland says that Yu Yangyi knew what he was doing but his opponent, Li Chao, did not. So in the video you will see Li Chao concentrating hard and continually looking at Yu's board to see what he played. Shankland was none too happy about this happening in such an important match.

Oh, interesting. I had a similar experience myself some years ago: personally I found it off-putting rather than helpful to have the same game going on at a nearby board.

[Chris Rice]

Emil Sutovsky was unhappy with the way Poland were paired during the event. I was surprised at this as Poland were always in the top places so they would get all the stronger countries wouldn't they? Here are his actual words from FB and let's bear in mind he was a key supporter of Dvorkovich:

"Again and again we speak about unfair and simply unsuitable pairing system in the Olympiad. This time Poland fell a victim - they had to play all eight highest-seed, whereas nobody else played more than four. The system was justly criticized in Tromso and Baku, but nothing changed. I promise to do my utmost to change it already next year. Talking and complaining not enough - time to act."

[Christopher Kreuzer]

Interesting tie-break analysis in the comments to the Chessbase quick response (and closing ceremony video) to the final results. I hope that "pd98" won't mind me reproducing what is said in full:

For those who are confused by the Sonneborn-Berger tiebreaker, I did some math out of curiosity, and here's how China overtake US in the final round.

The tiebreaker is the sum of the board points scored against a team multiplied by that team's match points in the tournament, without the team that has the lowest match point. At the end of round 10, US led China by 4 points http://chess-results.com/tnr368908.aspx ... 10&flag=30, and after round 11, China is up by 12 points.

So what made up the 16 point difference? The big one is Panama and Georgia3, they had the lowest match points (11) after round 10, and US scored 4:0 vs Panama (4x11=44) and 2.5:1.5 vs Georgia3 (2.5*11=27.5). Since we are dropping the team with the lowest match point, and given we have a tie, we dropped the 27.5 for US. But in round 11, Panama lost and Georgia3 won, so now instead of having 44 points between those two teams, US only got 2.5*13=32.5 for Georgia3. Lost of 11.5 points.

For China, the lowest opponent score was Morocco both after round 10 and 11, so nothing changed.

The rest of the opponents for US, Netherland/Israel/Bosnia/Croatia lost, India/Poland/Armenia draw (2.5+1.5+2.5)*1=6.5, Azerbaijan won 2.5*2=5, US gain 11.5 points.

The rest of the opponents for China, Colombia/Peru/Croatia/Netherland lost, Czech/Ukrain/Poland draw (1+2+3)*1=6, Azerbaijan/Iran won (2.5+2.5)*2=10, China gain 16 points.

Last time US won gold due to the same reason https://en.chessbase.com/post/tie-break ... s-olympiad, with Jordan (Ukrain beat them 4:0) dropped to the lowest opponent for Ukrain after last round, and let US won the gold. This time it's Panama (US beat them 4:0) dropped to the lowest opponent instead of Georgia3.

I wonder how much of this could have been calculated by the interested parties while the final round was in progress? The critical bit to me, seems to be the way the 'lowest scoring opponent' can change in the final round and have a massive impact on the final tie-break calculation. Without that, China would have still won, but only by half an SB point.

[Roger de Coverly]

Emil Sutovsky was unhappy with the way Poland were paired during the event. I was surprised at this as Poland were always in the top places so they would get all the stronger countries wouldn't they?

Isn't he complaining about the feature of Swiss systems that teams/players on the same score will meet? I think we've seen something similar in the British when it was eleven rounds and there was a sole leader who gets stronger opposition than the second placed challenger once they have already played.

For team events, they've already discarded pairing by game points and even discarded ranking by game points within match points. I'm not sure what else is left, other than fine tuning within the existing system. If you ranked within score group by average rating of the top four players, would that be any better?

[Christopher Kreuzer]

Having played through the game where Kosteniuk made an incorrect threefold repetition, I have some sympathy. She made the claim at move 83, and between moves 76 and 84 it does feel like the position might have repeated several times, but it hadn't been three times.

Note that the positions after White's 76th move and Black's 84th move are the same, but with a different player to move. Black effectively triangulated with her king on these squares: g6-f7-f6-f7-e8-d7-d6-e7-f6-g6. White had run out of checks by that point.

If Kosteniuk had made a second incorrect threefold repetition, would she have lost the game?

[JustinHorton]

No, you get two goes, at least according to the commentary at the time

[Roger de Coverly]

Having played through the game where Kosteniuk made an incorrect threefold repetition, I have some sympathy. She made the claim at move 83, and between moves 76 and 84 it does feel like the position might have repeated several times, but it hadn't been three times.

I'm marginally surprised it isn't a feature of the DGT boards (or rather the software that transmits the moves to the world) that they flag up repetitions, thus enabling arbiters to give an instant decision. But the arbiters at match level may not have access to this. "Computer says no" might not have been accepted by the player making the claim.

9.2.1 The game is drawn, upon a correct claim by a player having the move, when the same position for at least the third time (not necessarily by a repetition of moves):

They don't say how the claim should be validated. In the Kosteniuk game, the traditional method of playing through the game score seems to have been applied.

[JustinHorton]

Having played through the game where Kosteniuk made an incorrect threefold repetition, I have some sympathy.

Following it at the time, it was - to me - an absurd claim, because the king was on e7 and it had clearly not previously gone to that square since the lawn arrived on d2.

This seemed so obvious that I even wondered if she'd made the claim simply to buy herself a breathing space to think about the position. But I then put on the live video and it was instantly clear that this had been an unworthy thought, because you could see her looking up and down her scoresheet in the way somebody does when they know something ought to be absolutely clear, and that it probably is, but they're not getting it and they know they're not going to.

As you'd imagine, it was also clear that she was never going to save the game from there. Really painful stuff to watch.

The video's here and they go to Kosteniuk's game at about 5:19, though I started watching later than that.

[Roger de Coverly]

Interesting tie-break analysis in the comments to the Chessbase quick response (and closing ceremony video) to the final results.

That suggests it would be more stable if all the teams were included, rather than excluding the lowest placed. Teams of comparable ranking will play comparable opposition in the first round. Perhaps this would place more weight than desirable on first round bashing.

[Christopher Kreuzer]

I see what you mean about Kosteniuk's body language. There is some case to be made for admiring how Ju held it together to avoid moving her king to squares where Kosteniuk could have made a threefold repetition the following move. I think this happened twice. Can anyone say how many times the same position did actually get repeated twice and what those positions were? I am trying to do that, but struggling!

[Christopher Kreuzer]

Interesting tie-break analysis in the comments to the Chessbase quick response (and closing ceremony video) to the final results.

That suggests it would be more stable if all the teams were included, rather than excluding the lowest placed. Teams of comparable ranking will play comparable opposition in the first round. Perhaps this would place more weight than desirable on first round bashing.

Maybe instead of dropping the results against the lowest placed team, reduce it in some way (halve it?) or something like that?

[JustinHorton]

I am deeply sceptical that anything will come of the tie-break complaints, because to do that, you need to grasp the nettle that there is a trade-off between comprehensibility and fairness, and you have to decide which one of these you want to prioritise. You can tweak Sonneborn-Berger all you want, but if that's your preferred tie-break then you're stil lgoing to be following games on much lower boards a couple of hours after the main matches have finished. So you need to decide do we want to avoid that, yes or no? And I don't think they're going to do that.

[Christopher Kreuzer]

The video's here and they go to Kosteniuk's game at about 5:19, though I started watching later than that.

The threefold claim occurs at 5:29:50.