Olympiad

The very latest International round up of English news.
Andy Howie
Posts: 160
Joined: Wed Jul 15, 2009 3:32 pm

Re: Olympiad

Post by Andy Howie » Thu Aug 14, 2014 11:08 pm

I have just received a tweet that a second player has died today. The person is normally reliable with news. No more information as yet. I am hoping it is just someone misreading the Norwegian news as they were reporting an Asian had died earlier but I have a bad feeling about this

Roger de Coverly
Posts: 18662
Joined: Tue Apr 15, 2008 2:51 pm

Re: Olympiad

Post by Roger de Coverly » Thu Aug 14, 2014 11:19 pm

JustinHorton wrote:Right, this is killing me. I cannot work out how Hungary's tiebreak score of 372.0 has been arrived at. I've tried this more than a dozen times now and can't get to that figure.

I get this. Can any kind person tell me what I have missed or got wrong?

Algeria. 4. 9. Neglected as lowest-scoring opponent.
Venezuela. 2.5. 13. 32.5. 32.5.
The way it seems to work is that you do four nines = 36 and then two and half times thirteen = 32.5

So it's Venezuela that is disregarded, not Algeria. So the weak team dispatched 4-0 in the first round can improve the tie-break, just as long as you get the four points.

Whether that's the intended approach, or just the way they wrote the program is another matter.

You could also disregard the highest points, but that is probably undesirable from a sporting viewpoint. India's rampage in the last round took them to Bronze.

Some drastic consequences for the ranking of English players. David improves whilst Nigel drops to sixth place. When was he last that low? Presumably 1983 to 1984 as he made his debut as "second reserve" in 1984, if I remember correctly.

User avatar
JustinHorton
Posts: 7550
Joined: Mon Aug 04, 2008 10:06 am
Location: Somewhere you're not

Re: Olympiad

Post by JustinHorton » Thu Aug 14, 2014 11:25 pm

Roger de Coverly wrote:
JustinHorton wrote:Right, this is killing me. I cannot work out how Hungary's tiebreak score of 372.0 has been arrived at. I've tried this more than a dozen times now and can't get to that figure.

I get this. Can any kind person tell me what I have missed or got wrong?

Algeria. 4. 9. Neglected as lowest-scoring opponent.
Venezuela. 2.5. 13. 32.5. 32.5.
The way it seems to work is that you do four nines = 36 and then two and half times thirteen = 32.5

So it's Venezuela that is disregarded, not Algeria. So the weak team dispatched 4-0 in the first round can improve the tie-break, just as long as you get the four points.
I'm sure you're right, but in that case why not disregard China (one and a half times nineteen = 28.5) instead? Which would be silly, but I'm not seeing how they're doing what they're doing, even though it does indeed seem to be what they're doing.
"Do you play chess?"
"Yes, but I prefer a game with a better chance of cheating."

lostontime.blogspot.com

Roger de Coverly
Posts: 18662
Joined: Tue Apr 15, 2008 2:51 pm

Re: Olympiad

Post by Roger de Coverly » Thu Aug 14, 2014 11:34 pm

JustinHorton wrote: I'm sure you're right, but in that case why not disregard China (one and a half times nineteen = 28.5) instead? Which would be silly, but I'm not seeing how they're doing what they're doing, even though it does indeed seem to be what they're doing.
Easy enough to code, I would suspect. You select the list of potential discards only from the matches won (or drawn (?) ). There's logic to that as well, if you got thumped 4-0 by the eventual winners, should that be ignored? It's the ranking for pairings as well as tie-breaks, so it has to work all the way down to the bottom teams.

Andy Howie
Posts: 160
Joined: Wed Jul 15, 2009 3:32 pm

Re: Olympiad

Post by Andy Howie » Thu Aug 14, 2014 11:49 pm

Andy Howie wrote:I have just received a tweet that a second player has died today. The person is normally reliable with news. No more information as yet. I am hoping it is just someone misreading the Norwegian news as they were reporting an Asian had died earlier but I have a bad feeling about this
Just been told that a member of the ICCD team has died tonight. Sad news indeed

Lewis Martin
Posts: 296
Joined: Wed May 28, 2014 11:45 am

Re: Olympiad

Post by Lewis Martin » Fri Aug 15, 2014 12:32 am

Roger de Coverly wrote:
JustinHorton wrote:Right, this is killing me. I cannot work out how Hungary's tiebreak score of 372.0 has been arrived at. I've tried this more than a dozen times now and can't get to that figure.

I get this. Can any kind person tell me what I have missed or got wrong?

Algeria. 4. 9. Neglected as lowest-scoring opponent.
Venezuela. 2.5. 13. 32.5. 32.5.
The way it seems to work is that you do four nines = 36 and then two and half times thirteen = 32.5

So it's Venezuela that is disregarded, not Algeria. So the weak team dispatched 4-0 in the first round can improve the tie-break, just as long as you get the four points.
No, it is the country with lowest number of points, not the lowest Sonneborn-Berger. So, it is Algeria being neglected.

In fact, assuming that the other countries in that tie-break are correct (I have verified it for India), I have Hungary with 344.5 making them 5th! And therefore China, India and Russia being the medallists.

John McKenna
Posts: 4140
Joined: Tue May 17, 2011 2:02 pm

Re: Olympiad

Post by John McKenna » Fri Aug 15, 2014 1:29 am

Using all 11 such individual tie-break results for Hungary the total is 411.5.

Hungary has no tie-break result of 39.5 that would give the published T/B2 of 372.0 (411.5-39.5).

Eliminating the Algeria tie-break result (4*9=36) gives Justin's original T/B2 375.5 (411.5-36).

This implies that Justin's 375.5 is right and that there has been a mistake in the 372.0 official figure.

Very sorry to hear about the passing away of two players at the event.

Edit: Amended in the light of what Lewis Martin has posted below.
Last edited by John McKenna on Fri Aug 15, 2014 11:24 am, edited 1 time in total.
To find a for(u)m that accommodates the mess, that is the task of the artist now. (Samuel Beckett)

Lewis Martin
Posts: 296
Joined: Wed May 28, 2014 11:45 am

Re: Olympiad

Post by Lewis Martin » Fri Aug 15, 2014 9:41 am

Andy Howie wrote:
Andy Howie wrote:I have just received a tweet that a second player has died today. The person is normally reliable with news. No more information as yet. I am hoping it is just someone misreading the Norwegian news as they were reporting an Asian had died earlier but I have a bad feeling about this
Just been told that a member of the ICCD team has died tonight. Sad news indeed
I did not want to wish to comment on this until I knew more since the above suggests itself. Waking up this morning, a member of the ICCD team messaged about this.

He wasn't old, but whenever I have seen him, he always looked older than his age. But still, a great shame. From the little I know of him, he had the look of someone with steely eyes over the board, but off it he was always cheery.

Thinking of the fellow ICCD players and his family at this time.

Alex Holowczak
Posts: 9085
Joined: Sat May 30, 2009 5:18 pm
Location: Oldbury, Worcestershire
Contact:

Re: Olympiad

Post by Alex Holowczak » Fri Aug 15, 2014 1:03 pm

I've just investigated the tie-breaks.

I got India to be 371.5, having discarded the result against Syria, because they had the lowest number of matchpoints.

There is a complication in Hungary's tie-break calculation. I too got 375.5 the first time through. "Why is this 3.5 above their published score?" I saw that Hungary beat Portugal 3.5-0.5, and if you take a matchpoint off Portugal, you get the right score. I looked at the Portuguese results, and lo and behold: They beat Mali 4-0 by default in Round 1.

This situation is covered by this regulation:
Olympiad Regulations wrote:For tie-break purposes a bye or an unplayed match - if the opponent team does not appear on time – are counted as a drawn match against a virtual opponent. At the start of the round this virtual opponent has the same number of match points and game points as the team present. Then the result of the unplayed match is added (1 matchpoint and 4 gamepoints in case of an unplayed win, 1 matchpoint and 2 gamepoints in case of a bye, 0 matchpoints and 0 gamepoints in case of an unplayed loss) and finally for each subsequent round 1 matchpoint and 2 gamepoints.
Therefore, Portugal got 2 matchpoints for beating Mali, but any team that played Portugal in the tournament subsequently were only able to count 1 of those matchpoints towards their tie-break score. As a result, rather than doing 13*3.5 for Hungary v Portugal, you have to calculate 12*3.5. If you do that, and work out the calculation, Hungary get 372, rather than 375.5.

So Hungary's tie-break score was affected by Mali's non-appearance to play Portugal in Round 1. Had Mali appeared and lost 4-0, as would be expected, then Hungary's tie-break would have been 375.5. This might have been controversial if India's tie-break score was somewhere between 372 and 375.5!

User avatar
JustinHorton
Posts: 7550
Joined: Mon Aug 04, 2008 10:06 am
Location: Somewhere you're not

Re: Olympiad

Post by JustinHorton » Fri Aug 15, 2014 1:09 pm

Thanks very much. Obviously I noticed I was 3.5 off and therefore, seeing Hungary had beaten Portugal 3.5-0.5, the Portugal score was the first thing I checked, but I didn't know about the default.

I assume this disproves Roger's theory?

Anybody know why Mali defaulted?
"Do you play chess?"
"Yes, but I prefer a game with a better chance of cheating."

lostontime.blogspot.com

Alex Holowczak
Posts: 9085
Joined: Sat May 30, 2009 5:18 pm
Location: Oldbury, Worcestershire
Contact:

Re: Olympiad

Post by Alex Holowczak » Fri Aug 15, 2014 1:15 pm

JustinHorton wrote:Thanks very much. Obviously I noticed I was 3.5 off and therefore, seeing Hungary had beaten Portugal 3.5-0.5, the Portugal score was the first thing I checked, but I didn't know about the default.
Neither did I until I looked for it!
JustinHorton wrote:I assume this disproves Roger's theory?
The regulations are clear:
Olympiad regulations wrote:match points of each opponent, excluding the opponent who scored the lowest number of match points, multiplied by the number of game points achieved against this opponent
Take a look here to read them in full: http://www.fide.com/fide/handbook.html? ... ew=article
JustinHorton wrote:Anybody know why Mali defaulted?
Mali were not paired in any subsequent round, so I think the correct question to ask is "Anybody know why Mali were paired in Round 1?" Speculation on my part would be that they never made it.

Alistair Campbell
Posts: 311
Joined: Sat Mar 06, 2010 12:53 pm

Re: Olympiad

Post by Alistair Campbell » Fri Aug 15, 2014 2:18 pm

Alex Holowczak wrote:I've just investigated the tie-breaks.
Thanks - that has set my mind at rest. I couldn't remember having any particular difficulty calculating such tie-breaks at the European Club Cup last year (they were crucial in determining who had the bye) so I was slightly worried when I could only muster a batting average of 1/5 from my attempted calculations this time round.

Foolishly, I had thought that calculating tie-breaks for teams with few match points (such as Rwanda and the Solomon Islands) might have reduced the possibility of error, but in fact, due to the number of defaults at that end of the draw, it enhanced it.

My calculation for Scotland's Open team now matches as both Zimbabawe and Monaco had managed a win by default, which means I must subtract 5.5 from my total (having eliminated Tanzania's contribution). So all in all, a constructive morning...

But what happens if there is more than one team with the lowest number of match points - which do you eliminate then? Presumably this is based on notional match points, rather than actual match points?

Roger de Coverly
Posts: 18662
Joined: Tue Apr 15, 2008 2:51 pm

Re: Olympiad

Post by Roger de Coverly » Fri Aug 15, 2014 2:27 pm

A wrong type of air excuse for the relatively poor performance by the top board of the hosts, who unlike their opposition were on the same boards for the whole tournament. But the Chinese were in the same area for at least the second half.

http://www.nrk.no/sport/tung-luft-kan-h ... 1.11878666

The Google translate comes up as "heavy air", but the natural translation from the context is poor or stale air.

Alex Holowczak
Posts: 9085
Joined: Sat May 30, 2009 5:18 pm
Location: Oldbury, Worcestershire
Contact:

Re: Olympiad

Post by Alex Holowczak » Fri Aug 15, 2014 3:27 pm

Alistair Campbell wrote:Thanks - that has set my mind at rest. I couldn't remember having any particular difficulty calculating such tie-breaks at the European Club Cup last year (they were crucial in determining who had the bye) so I was slightly worried when I could only muster a batting average of 1/5 from my attempted calculations this time round.
Were they using the same tie-break rules? I know they use a different pairing system.
Alistair Campbell wrote:But what happens if there is more than one team with the lowest number of match points - which do you eliminate then? Presumably this is based on notional match points, rather than actual match points?
Don't know. Might be worth finding a practical example and working out what they did.

John Moore
Posts: 1870
Joined: Wed May 14, 2008 6:33 pm

Re: Olympiad

Post by John Moore » Fri Aug 15, 2014 4:52 pm

I find it incredibly difficult to get excited about how the tiebreaks work but there are people who seem to spend an awful lot of time on it.

Post Reply