PDA

View Full Version : NSWCA Ryde Eastwood Weekender



Thunderspirit
13-10-2004, 02:04 PM
Hi Everyone,
This link is to make a public apology to all the players who played at Ryde Eastwood Weekender during the long weekend. I had a very poor weekend as a DOP, and many mistakes I am still very unhappy with. My decision not to change the round 6 pairings was the worst of quite a few errors, with many players on the top boards being effected.

Special apologies must extend to FM Greg Canfell, Brian Fitzpatrick and Milan Grcic who worse the worst effected.

It was a learning experience, and I will endevaour to do a better job next time.

arosar
13-10-2004, 02:25 PM
This link is to make a public apology . . . .

Yo Leeeee mann!!! How's things dude?

We forgive for your DOP mistakes man. But we cannot forgive you for posting the thread in the wrong forum. :)

AR

antichrist
13-10-2004, 05:53 PM
Hi Everyone,
This link is to make a public apology to all the players who played at Ryde Eastwood Weekender during the long weekend. I had a very poor weekend as a DOP, and many mistakes I am still very unhappy with. My decision not to change the round 6 pairings was the worst of quite a few errors, with many players on the top boards being effected.

Special apologies must extend to FM Greg Canfell, Brian Fitzpatrick and Milan Grcic who worse the worst effected.

It was a learning experience, and I will endevaour to do a better job next time.

Lee,
Was the draw done by computer (of course)?

Bill Gletsos
13-10-2004, 06:31 PM
Lee,
Was the draw done by computer (of course)?
Yes it was.

Thunderspirit
13-10-2004, 08:41 PM
To answer Peter's question,
the pairings were done with Swiss Perfect, but I didn't check them before making them public, and George Xie had rung me to ask for the pairings, which I gave him. In the morning I was informed that the pairings were wrong (which they were) but the dillemma was: do I change them, though they were wrong. I decided not to change them, which was a mistake.

jenni
13-10-2004, 09:26 PM
So it was the computer doing them wrong - rather than you? One of the bad pairings that SP does from time to time, in its pursuit of colour balance?

Ian_Rogers
13-10-2004, 09:38 PM
Hi there. Cathy here.

I wasn't at the event which Lee arbited so I don't know how bad his mistakes really were. However, I would like to congratulate Lee for his public apology. I think one of the most important qualities to make a good arbiter is the ability to admit your mistakes rather than brazening it out and trying to justify bad decisions.
Cathy

Kevin Bonham
13-10-2004, 11:06 PM
To answer Peter's question, the pairings were done with Swiss Perfect, but I didn't check them before making them public, and George Xie had rung me to ask for the pairings, which I gave him. In the morning I was informed that the pairings were wrong (which they were) but the dillemma was: do I change them, though they were wrong. I decided not to change them, which was a mistake.

Actually all you did wrong was make the draw public before it had been thoroughly checked. I had this exact problem a few years ago and there is actually a pairing regulation which says:

F6. A pairing officially made public shall not be changed unless it violates the absolute pairing criteria (B1 and B2).

I think that if you tell a player they are definitely playing another then that's equivalent to making the draw public, and you shouldn't change the draw after that unless you can do so without changing that board.

(B1 and B2, when not chasing teddybears, state:

B1(a) two players shall not meet more than once,
(b) a player who has received a point on bye or forfeit shall not receive a bye
B2(a) no player's colour difference will be >+2 or <-2
(b) that no player will get the same colour three times in a row. )

The case where I made the same mistake of declaring a computer pairing final when it was wrong was rather nasty as it upfloated a midfield player to play one of the co-leaders who thereby finished =1st in the event. We did implore him to accept a change of draw but he stuck to his guns (as he was entitled to do) and we had to go with the defective draw because of rule F6.

We have since dealt with F6 by declaring that in all TCA events all swiss draws made public are considered provisional and unofficial draws only until 10 minutes before the next round, and the DOP cannot override this declaration. Crude but effective.

Ian Rout
14-10-2004, 08:56 AM
I think it's a good habit to always say when advising someone of their expected next round opponent that it is not definite. You can either just say that or explain why in detail, depending how much spare time you have. To commit to a draw is unnecessarily restricting yourself, especially where it allows someone to lock in a relatively soft game.

The alternative is to refuse to answer. I think this is not a good look - it's officious and even childish ("I've got a secret and I won't tell"), and also gives an advantage to those who know the pairing rules and can work it out themselves.

In ACT club tournaments the next draw is normally circulated by e-mail as a matter of course without people asking, but as Canberra is a city of public servants who are aware of F6, always stating the proviso that it is provisional and subject to change.

jenni
14-10-2004, 09:37 AM
Hi there. Cathy here.

I wasn't at the event which Lee arbited so I don't know how bad his mistakes really were. However, I would like to congratulate Lee for his public apology. I think one of the most important qualities to make a good arbiter is the ability to admit your mistakes rather than brazening it out and trying to justify bad decisions.
Cathy

I agree with this - I think Lee is trying very hard to be a good DOP and shouldn't worry about it too much. He was very depressed about the tournament on the Monday and as I said to him then, you can't keep everyone happy. He had got himself into a situation where he was going to annoy someone and sometimes you just have to live with it. It was all good experience and he will now have strategies to deal with a similar situation.

There were a number of very little errors, the major one was the pairing problem. I think Gareth might have been affected, as I was told he shouldn't have been playing Michael. However Gareth didn't care - he likes playing Michael. Gareth won their last encounter, so it was time for Michael to exact his revenge. :)

Thunderspirit
14-10-2004, 10:25 AM
Hi Everyone,
Just to clarify a few points. The pairings that were made didn't violate the 2 fundimental pairing rules, being that players can't be paired twice or recieve 3 of the same colour in a row (when the last round isn't the last round).
On the pairings and Swiss Perfect. Swiss Perfect is just a tool, it's the DOP who does the pairings.
Finally I'd like to thank Jenni, who spent much on Monday trying to cheer me up, and Cathy whose advice as a long time IA is always useful.

Bill Gletsos
14-10-2004, 11:47 AM
Personally I dont think Lee should be too hard on himself.
In fact having had a close look at the SP files I dont think the Swiss Perfect draw for round 6 was actually incorrect.

After 5 rounds Xie was on 5 points with an absolute color preference of black due to having had white in the previous two rounds.
There were 9 players following on 4 points.
Although Greg Canfell had an absolute color preference of white due to having had black in the previous two rounds he also had upfloated in round 4 when with 2.5 points he played Junta Ikeda on 3 points.
Bolens the next highest available player also had an absolue color preference of white, however Bolens had upfloated in the 5th round when he on 3 points played Alex Mendes da Costa on 3.5.

The FIDE rules will not allow for the pairing of Xie against Canfell or Bolens unless absolutely necessary under the rules (mainly due to other pairings of players on 4 points violating B1 or B2). As such the pairings generated by SP would in fact be the same pairings generated by an arbiter doing the pairings manually if they followed the FIDE rules.

Based on my reading and understanding of the FIDE Swiss Rules although the pairings generated by Sp may appear wrong to the casual observer they are in fact correct according to the FIDE rules.

Lucena
14-10-2004, 01:25 PM
Personally I dont think Lee should be too hard on himself.
In fact having had a close look at the SP files I dont think the Swiss Perfect draw for round 6 was actually incorrect.


Lee sounds pretty sure Bill. Can someone tell us what Charles Z thinks about this? Could he not have perhaps been the person to inform Lee that the draw was wrong in the first place?

Garvinator
14-10-2004, 01:30 PM
Lee, Bill someone, can you send me the sp files for this tournament at ggrayggray@hotmail.com.

Cheers,

Garvin Gray
2004 Grand Prix Co- Ordinator.

Bill Gletsos
14-10-2004, 01:38 PM
Lee sounds pretty sure Bill.
I had spoken to Lee on the phone about it my view.
He had not as far as I can tell actually tried doing the draw manually but had just accepted the conventional view espoused by some that the draw was wrong given at a glance that Canfell could have played Xie.
Based on the FIDE Swiss rules this belief is actually incorrect.


Can someone tell us what Charles Z thinks about this? Could he not have perhaps been the person to inform Lee that the draw was wrong in the first place?
I dont think Charles has offered an opinion but I believe he would agree with me.

As I said I believe Lee just took some peoples belief that the draw must be wrong as a given because on appearance that appears to be the case.

I'm sure Lee will clarify the above when next he posts.

Bill Gletsos
14-10-2004, 01:40 PM
Lee, Bill someone, can you send me the sp files for this tournament at ggrayggray@hotmail.com.

Cheers,

Garvin Gray
2004 Grand Prix Co- Ordinator.
Lee doesnt have them as the event wasnt run on his computer.
I'll send them to you when I get home tonight.

Thunderspirit
14-10-2004, 02:08 PM
Hi All (Again!!),
Bill is right on the fact the pairings being correct strictly under the rules. Technically a DOP shouldn't give the same player two indentical floats two rounds apart. In English that is player 'x' shouldn't have an upfloat in Rounds 3+5. Though this is strictly correct, I know quite a few DOP's including IA's who don't strictly enforce this. If I would have known I would have paired Greg vs George in R 6 because it is the common sense pairing.

Thanks to Bill, for ringing me, and Gareth for sticking up for me.

As for Galvin wanting the SP files, my own personal opinion is a DOP's work should n't just handed around willy nilly. Bill needs the files to do the ratings, and should n't be given to third parties without a vaild reason.

Garvinator
14-10-2004, 02:16 PM
Bill needs the files to do the ratings, and should n't be given to third parties without a vaild reason.
Lee, sending the files to me is not handing them out willy nilly. I require them as Grand Prix Co-ordinator as it makes the task of identifying juniors and women players alot easier. The NSWCA Ryde Eastwood Weekender is a category 3 Grand Prix tournament. I recieve the sp files from almost all grand prix tournaments.

Cheers,

Garvin Gray
2004 Grand Prix Co-ordinator

Bill Gletsos
14-10-2004, 02:35 PM
Hi All (Again!!),
Bill is right on the fact the pairings being correct strictly under the rules. Technically a DOP shouldn't give the same player two indentical floats two rounds apart. In English that is player 'x' shouldn't have an upfloat in Rounds 3+5. Though this is strictly correct, I know quite a few DOP's including IA's who don't strictly enforce this. If I would have known I would have paired Greg vs George in R 6 because it is the common sense pairing.

Provided the round draw genertated by the SP program is correctly following the FIDE pairing rules I dont think it is at all wise to override it even if doing so may appear to be a case of apparent common sense.

I think the danger in such a situation is that it opens the arbiter up to claims of unfair and biased manipulation of the draw especially since the arbiters pairing is now one of the arbiter not following the FIDE swiss pairing rules. Given the arbiter in such circumstances cannot reasonably argue the FIDE rules are ambiguous or open to interpretation then the arbiter has no valid defence with regards to his actions. Such manual changes could also cause an argument at the tournament site by a player who actually knows the FIDE swiss rules (given few even know the laws of chess then the chnaces of this happening in reality would be slight :) ) disputing the draw with the arbiter.

As such unless the arbiter can prove that the computer generated draw is incorrect according to the rules he should not override it and make manual changes.

Denis_Jessop
14-10-2004, 05:25 PM
Based on my reading and understanding of the FIDE Swiss Rules although the pairings generated by Sp may appear wrong to the casual observer they are in fact correct according to the FIDE rules.

I had nothing to do with this tournament but, speaking from my experience with SP and of some rather odd-looking pairings it produced in the last two rounds of the recent ACT Championship which I directed, I agree with Bill that often the pairings done by SP are correct according to the FIDE Swiss Rules and arbiters ought to be very wary of changing them. Also, the odd pairings are often brought about by application of the floating rules and this won't be apparent to anyone, including the players and the Arbiter, unless they have carefully studied the details of the tournament.

Regarding Jenni's reference to colour preference, it's not SP that is to blame here but the FIDE Swiss Rules, if blame is the right word. Moreover, as I have mentioned, it's just as likely to have been the floating rules that causes odd-looking pairings.

What I have found to be most curious is when SP downfloats a player across a score group for reasons that are most obscure. It did that to me in the ACT Championship in a case where the downfloater could quite legally have been paired against a player in the score group immediately below. It also managed to pair 3 players in one score group against 3 from the group below when there seemed no reason for some of the players in each group to be paired against players on the same score. For good measure, it gave the bye to the wrong player.

I suspect that the real problem is that the FIDE Swiss Rules are written with fairly high-powered tournaments in mind and they don't work so well when an event is, say, a small club event with a wide range of playing strengths.

So the moral is, Lee, don't fret - it happens to all of us and congratulations for being so open about it!

Denis Jessop

Garvinator
14-10-2004, 05:45 PM
What I have found to be most curious is when SP downfloats a player across a score group for reasons that are most obscure. It did that to me in the ACT Championship in a case where the downfloater could quite legally have been paired against a player in the score group immediately below. It also managed to pair 3 players in one score group against 3 from the group below when there seemed no reason for some of the players in each group to be paired against players on the same score. For good measure, it gave the bye to the wrong player.
were the players re paired manually in this situation? If not, why not?

jenni
14-10-2004, 06:13 PM
arbiters ought to be very wary of changing them.
Denis Jessop

This is interesting, as I have always got the impression that Charles Z always checks the pairings and changes them if he doesn't think they are right?

We have a policy in the ACTJCL of never changing a pairing, for two reasons.

1. most of us have no idea of what we are doing and even where some of us know a little, it is a case of a little knowledge is dangerous, so safer to leave alone even when the pairing looks odd.

2. particularly with juniors you have to have a perception of fairness. Often the person acting as DOP has a child playing and the other kids need to be sure that everything is being done correctly. Often one of the kids will do the pairing and then print the pairings, so that they know that no hidden manipulation is being done.

arosar
14-10-2004, 06:44 PM
So the moral is, Lee, don't fret - it happens to all of us and congratulations for being so open about it!

Yeah that's right. Quite frankly I don't even know what he's apologising about. I wasn't at the tourn - lemme say that. But I've been in tourns where Lee's officiated and I've always found him to be highly professional and very courteous towards players.

AR

Kevin Bonham
14-10-2004, 08:04 PM
What I have found to be most curious is when SP downfloats a player across a score group for reasons that are most obscure. It did that to me in the ACT Championship in a case where the downfloater could quite legally have been paired against a player in the score group immediately below. It also managed to pair 3 players in one score group against 3 from the group below when there seemed no reason for some of the players in each group to be paired against players on the same score.

Hard to comment on that one without seeing the files. A case I have seen (Aus Champs 2002) involved SP merging a scoregroup with 4 players in it with scoregroups below that even though two of the four players could clearly play each other. That was an instance of conflict between B3 and Section C where SP, in my view (although not all arbiters agree) incorrectly, gives section C priority over section B.

I have sometimes encountered very odd SP draws and sat down to work out whether they are correct implementations of the system. In around 90% of those cases I've checked the SP draw has turned out to be correct under the FIDE rules even when the draw looks really odd. In around the other 10% it is wrong. Many of these relate to the C/B conflict mentioned above.

I'll actually defend the draw produced by the computer and the FIDE rules in Lee's instance. The FIDE Dutch system (commonly known as "the swiss system") rules pair a player with their performance on the weekend in question the paramount consideration, and rating as a kind of tiebreak. If you don't like that, just don't use Dutch pairings at all; use something else, like the Crouch system. In Dutch pairings, if you don't have that bar on upfloats two rounds apart then the same high-rated player who is performing a bit below expectation just keeps getting upfloated and upfloated and upfloated every second round, just as (more concerningly) an overperforming low-rated player gets downfloated over and over (great way to get a ratings prize). The aim is to give a player a fair go - if you get upfloated then you deserve to not get upfloated when there is a chance for this to happen later.

Lee deserves a pat on the back for being so concerned about the tournament being a success that he's willing to accept responsibility for a problem that arose. However based on what I've now seen, it was not Lee's fault, not SP's fault, not FIDE's fault - just a case of a draw that made people go "that's not right" when actually there is a kind of logic behind it. I would not have even considered overriding the draw in question.

Thunderspirit
14-10-2004, 09:21 PM
Hi Everyone,
I don't have the SP files but Bill is probably right on the floats. I know some DOP's including IA's will float up one round apart (say Rounds 3 + 5) and I also believe in doing this too...

Denis_Jessop
14-10-2004, 10:07 PM
were the players re paired manually in this situation? If not, why not?
The short answer is "yes" .

Denis J

Denis_Jessop
14-10-2004, 11:05 PM
This is interesting, as I have always got the impression that Charles Z always checks the pairings and changes them if he doesn't think they are right?

We have a policy in the ACTJCL of never changing a pairing, for two reasons.

1. most of us have no idea of what we are doing and even where some of us know a little, it is a case of a little knowledge is dangerous, so safer to leave alone even when the pairing looks odd.

2. particularly with juniors you have to have a perception of fairness. Often the person acting as DOP has a child playing and the other kids need to be sure that everything is being done correctly. Often one of the kids will do the pairing and then print the pairings, so that they know that no hidden manipulation is being done.

I, too, am prepared to change a pairing if it appears that it is not in conformity with the Swiss rules. But I do this rarely as I agree with Kevin that, in my experience, about 90% (or perhaps even more) of the odd-looking ones are right. The beauty of SP is that, not only does it relieve arbiters of a mountain of work and allow one to print out the draw but, as Bill has mentioned, also relieves them of claims of bias. All that I get now (from one or two players) is "that looks wrong - in Sydney they always change the pairings in cases like this!!" which I treat with an appropriate response. A complicating factor, alluded to by Kevin, is that the Dutch system rules are not 100% cast iron in their drafting and are thus open to interpretation so that it does not necessarily follow that the writer of the SP program has correctly interpreted the rules.

Where SP is being used by arbiters who aren't completely on top of the Swiss rules, it is courting trouble to change pairings. Likewise, in many ACTJCL events, I imagine that there are often quite a lot of unrated players. The Dutch system which SP uses is described as a Swiss System Based on Rating. Using it with a lot of unrated players affects one of the key pairing criteria and so it is probably best to adopt the policy that Jenni mentions. Moreover changing pairings, in my view, doesn't just involve "fixing up" one or two but rather checking the whole draw or, at least, a substantial part of it which can be very time-consuming. An alternative could be to use the Elementary FIDE Swiss System described by Stewart Reuben but even that needs an experienced practitioner to make a draw in a reasonable time and I wouldn't advocate it to replace SP.

On another tack, a real plus for SP is its ease of use. I have looked at one or two other programs and have found them to be much more complex, though without a fair trial, I must confess. For example, recently Geurt Gijssen drew attention to the new Windows version of the Royal Dutch Chess Federation's program Swiss Master 5 which I found much less convenient than SP when I tried to run it recently alongside SP. perhaps this is a subject for another thread but I'd be interested to know if anyone else has given it or the FIDE program a try. If anyone can get the Dutch system right, it should be the Dutchmen!

Denis J

Bill Gletsos
15-10-2004, 12:06 AM
Hi Everyone,
I don't have the SP files but Bill is probably right on the floats. I know some DOP's including IA's will float up one round apart (say Rounds 3 + 5) and I also believe in doing this too...
The fact us that those doing this cannot justify it under the rules.
You are either playing by the FIDE Swiss System Dutch Rules or you are not.
They should not bastardise the rules just beacuse they dont like them or think their way is better.

The arbiters should only override SP when the pairings it generates dont follow the FIDE Swiss rules.

To ensure uniformity within our events then when it comes to the NSWCA I suspect in future we will direct our arbiters to not overide SP except where it can be shown that its pairing does not obey the FIDE rules.

Ian_Rogers
15-10-2004, 07:40 AM
The rule about not floating for two rounds after a float is not an obligatory part of the Dutch Swiss rules (unlike, say, two players not meeting twice) and many arbiters make an announcement at the start of a tournament that this clause in the rules will not be applied. (I believe there is one other clause in the Dutch rules which is semi-optional.)
From a personal viewpoint, the double float rule, when applied, has ruined a number of tournaments in which I have competed, not allowing the leader to meet the top seeded pursuer in a late round because the pursuer had floated up two rounds earlier.
It's biggest downside was shown in an Australian Open about a decade ago. Reilly and Johansen were leading clearly but in the final rounds only one of them got to play the top ranked challengers, because after that the double float prevented the other player from meeting the same strong players encountered by his rival.

Ian

Thunderspirit
15-10-2004, 08:08 AM
Hi Galvin,
I was unaware that you are GP coorinator. Seems like a good enough reason for me... :P

pax
15-10-2004, 09:44 AM
A valuable feature in a program like SP would be an "arbiter's report" on each round pairing, which indicates which rules have been used.

For example, players which must play a certain colour may be highlighted in some way, players that are forbidden to float likewise and so on. This sort of report might help arbiters interpret the draw without analysing everything from scratch, and help soothe disputes.

Pax

Garvinator
15-10-2004, 11:22 AM
Hi Galvin,
I was unaware that you are GP coorinator. Seems like a good enough reason for me... :P
i think alot of ppl are unaware there is a gp co ordinator and for those that are aware, are not aware that i am gp co ordinator ;) and its Garvin, not Galvin, who is galvin :lol:

Bill Gletsos
15-10-2004, 11:54 AM
The rule about not floating for two rounds after a float is not an obligatory part of the Dutch Swiss rules (unlike, say, two players not meeting twice) and many arbiters make an announcement at the start of a tournament that this clause in the rules will not be applied. (I believe there is one other clause in the Dutch rules which is semi-optional.)
Ian my understanding is that as far as the Dutch rules are concerned B6 (the rule in question) is not optional as far as whether the arbiter can just choose to ignore it or not but that its applicability is dependant on how far thru part C of the Swiss rules you get. e.g. It is not until part C.9 where it says: "Drop criterion B6 and B5 (in this order) for downfloats and restart at C4." that the arbiter can ignore applying B6.

I ama aware that the FIDE Swiss rules are ambiguous and even apparently incorrect in some parts.e.g. In C1 and C6 where it mentions B1 & B2 apparently B3 to b6 are still applicable. In fact in an attempt to rectify and clarify this situatiion Stewart Reuben in his 2nd edition of his Chess Organisers Handbook renumbered B1 to B6 as a new B1 & B2 where the new B1 was the old B1 & B2 and his new B2 was the old B3 to B6. However in doing this he failed to renumber references to B5 & B6 instead of his new numbering.
It is no wonder this whole area is far from easy if not impossible for the average player to follow.

Unfortunately the FIDE Swiss pairing committee has not seen fit to clarify any of this in the 10 odd years that the Dutch rules have been in place.

Can you or Cathy add any further light on any of this.

Denis_Jessop
15-10-2004, 05:53 PM
I ama aware that the FIDE Swiss rules are ambiguous and even apparently incorrect in some parts.e.g. In C1 and C6 where it mentions B1 & B2 apparently B3 to b6 are still applicable.

Isn't there a further complication as far as using SP is concerned? I understand that the author of the SP program is not prepared to change it or to authorise anyone else to do so, so that, if the FIDE Dutch Rules are changed or clarified to have a meaning different from that accorded by SP, we are stymied. Hence my query about experience with other programs.

Denis J

Garvinator
15-10-2004, 05:55 PM
Isn't there a further complication as far as using SP is concerned? I understand that the author of the SP program is not prepared to change it or to authorise anyone else to do so, so that, if the FIDE Dutch Rules are changed or clarified to have a meaning different from that accorded by SP, we are stymied. Hence my query about experience with other programs.

Denis J
about the only other pairing program i am aware of is Swiss Sis, which Norm Braybrooke uses at Redcliffe Chess Club.

Denis_Jessop
15-10-2004, 06:10 PM
about the only other pairing program i am aware of is Swiss Sis, which Norm Braybrooke uses at Redcliffe Chess Club.

Recently the Royal Dutch Chess Federation put out a Windows version of Swiss Master 5, apparently only available in DOS format before. Geurt Gijssen seems to think it is great. Also FIDE put out a pairing program a couple of years ago but I can't trace it quickly - it may have only been a beta version anyway. I have found SM5 pretty cumbersome, but only on a very brief trial I should add.

Denis J

arosar
15-10-2004, 06:16 PM
Also FIDE put out a pairing program a couple of years ago but I can't trace it quickly - it may have only been a beta version anyway.

That's right. In fact I played around with the FIDE app. It's on my old Win95 box. Let me see if I can get it.

AR

Bill Gletsos
15-10-2004, 07:08 PM
Isn't there a further complication as far as using SP is concerned? I understand that the author of the SP program is not prepared to change it or to authorise anyone else to do so, so that, if the FIDE Dutch Rules are changed or clarified to have a meaning different from that accorded by SP, we are stymied. Hence my query about experience with other programs.

Denis J
That is my understanding as well.

The author of SP has developed a new program called League Watch which is currently in beta and available for test from the www.swissperfect.com

The League Watch program can import Swiss Perfect files and convert them to League Watch format files.

Of course in the ACF Bulletin #47 Graeme Gardiner said in regards to Swiss Perfect and its author: "Most importantly, he has demonstrated that he is prepared to make additions/changes to the programme at relatively short notice and that he is committed to developing the programme for the benefit of Australian chess."

That bulletin was dated 12th December 1999. The last update to SP98 occurred in February 2000. As part of the developemnt of a new rating system (Glicko) which debuted in Dec 2000 we made the rating system use SP files as its source of tournament results. Following this introduction I had a number of discussions with the author who at the time implied he would be prepared to implement some features. Unfortunately the implementation of these never eventuated. On top of that a number of long standing bugs in SP have never been fixed.

Now we have League Watch. I know neither Graham Saint nor I have been approached to provide any input with regards desirable features for the rating system.

I would be recommending to the ACF that before it considers investing in supporting League Watch as a recommended program like it did with Swiss Perfect that at the very least it would want some form of commitment that any bugs reported would be fixed in a quick and timely manner.


At this time there is no mention on the web site what the cost of League Watch will be and no mention if users of Swiss Perfect can upgrade to League Watch for some sort of "upgrade" fee as opposed to paying a full fee.

Thunderspirit
15-10-2004, 09:59 PM
Yeah that's right. Quite frankly I don't even know what he's apologising about. I wasn't at the tourn - lemme say that. But I've been in tourns where Lee's officiated and I've always found him to be highly professional and very courteous towards players.

AR

Hi Everyone,
These are kind words by 'Arosar' but sadly I don't know who you are!

Bill Gletsos
15-10-2004, 10:25 PM
Hi Everyone,
These are kind words by 'Arosar' but sadly I don't know who you are!
Thats Amiel.

Kevin Bonham
16-10-2004, 01:57 AM
The rule about not floating for two rounds after a float is not an obligatory part of the Dutch Swiss rules (unlike, say, two players not meeting twice) and many arbiters make an announcement at the start of a tournament that this clause in the rules will not be applied. (I believe there is one other clause in the Dutch rules which is semi-optional.)

This may help with Bill's question above:

There is nothing in section C.04.1 but at the head of Section C.04.2 "Regulations for FIDE Swiss System Tournaments" I noticed the following:


Scope: These regulations are to be used in FIDE competitions and in FIDE registered competitions which are declared to be conducted by "FIDE Swiss Rules". In this case, only minor departures from these regulations are permitted, and such departures must be declared before the competition begins and the attention of participants specially drawn to the departures.

I was going to post something to this effect re Aus tournaments - that minor (or even quite substantial) variations should be considered fine so long as they are declared in advance. If they are not declared in advance then the arbiter making an exception is open to claims of selectivity or bias.

Thinking about this a bit more, I think the arbiter should probably announce in advance that the double-float two rounds apart rule will be overridden if all the following are true:

(a) the arbiter feels comfortable and capable fixing such draws manually as needed
(b) the event is dominated (in ratings terms) by a small group of entrants (as sometimes happens when a few players are rated hundreds of points above everyone else)
(c) the number of rounds is relatively small for the number of entrants.

Ian_Rogers
16-10-2004, 07:12 AM
Hello there. Cathy here.

I think Kevin's post covers Bill's question to Ian and me very well. I can see absolutely no problem with announcing at the start of the tournament that you are going to do away with the double float rule.
Indeed I think the rule that you can't float again in the same direction as you floated two rounds before to be an absolutely awful rule and it should in my opinion be done away with completely.
The rule about not floating in the same direction in two consecutive rounds can also cause huge problems and I think an appropriate rule change which could be announced at the start of a tournament could be that you WILL float a player twice unless a floater within 100 rating points can be exchanged.

Ian here, now.

I don't think Kevin's precondition for dropping the B6 rule - that the tournament be of a short length - relevant. The problems in the Australian Open I mentioned earlier happened in the latter rounds of an 11 round event with a relatively modest number of participants.

Thunderspirit
16-10-2004, 10:51 AM
Hi Everyone,
I agree that for most Australian events, especially non FIDE rated events, the double float rule should not apply. The only problem with annoucing such a rule change before an event is that 90% of the players have probably never heard of the rule, and then the DOP will have to spend the next four rounds trying to explain it.
The key is to have a NSWCA policy on pairings, so if the DOP needs to change the pairings there is always a policy to support the DOP. When I spoke to Bill on the Ph, we did discuss this, so it is a possibilty.

Bill Gletsos
16-10-2004, 04:29 PM
This may help with Bill's question above:

There is nothing in section C.04.1 but at the head of Section C.04.2 "Regulations for FIDE Swiss System Tournaments" I noticed the following:
C.04.1 is the Dutch System.
C.04.2 describes a generic Swiss system which FIDE refers to as the FIDE Swiss Rules.
C.04.3 is the Dubov System

I would suggest that the C.04.2 only applies to the pairing rules described under C.04.2 and not to the Dutch system under C.04.1 nor the Dubov system under C.04.3

Bill Gletsos
16-10-2004, 04:37 PM
Hello there. Cathy here.

I think Kevin's post covers Bill's question to Ian and me very well. I can see absolutely no problem with announcing at the start of the tournament that you are going to do away with the double float rule.
Indeed I think the rule that you can't float again in the same direction as you floated two rounds before to be an absolutely awful rule and it should in my opinion be done away with completely.
Dont get me wrong Cathy, I'm not defending rule B6 just the application and interpretation of the Dutch rules as written.
When it comes to B6 I agree with you and think it should be dropped entirely.


The rule about not floating in the same direction in two consecutive rounds can also cause huge problems and I think an appropriate rule change which could be announced at the start of a tournament could be that you WILL float a player twice unless a floater within 100 rating points can be exchanged.
I agree with you here too.

I notice in the draft of the FIDE Handbook for 2005 on the FIDE web site the whole section regarding Swiss rules etc is missing and says it will be determined at the Olympiad. This could hopefully imply that changes are afoot otherwise there would seem to be no reason why they just didnt insert the old rules there immediately.


Ian here, now.

I don't think Kevin's precondition for dropping the B6 rule - that the tournament be of a short length - relevant. The problems in the Australian Open I mentioned earlier happened in the latter rounds of an 11 round event with a relatively modest number of participants.
I agree Ian.

Thunderspirit
18-10-2004, 11:31 AM
Charles Z here from Lee's account. A few points:

(a) I was not the one who told Lee the Round 6 pairings were wrong (I left the weekender on the Sunday at lunchtime, so was not aware of the problem until after the tournament), but agree with Bill (it happens occasionally!) that the pairings look like they were actually technically correct.

(b) Pairings should be provisional for weekenders, but made public (and not changed except under extreme circumstances) for once a week events like the NSW State Champs so players can prepare.

(c) Computers are a TOOL to aid the DOP, not the be-all and end-all, so pairings should be changed if they're wrong (as I feel Swiss Perfect often is).

(d) I would change the float rules by omitting B2(d) (cannot float in the same direction as two rounds before), but keep B2(c) (cannot float in the same direction as the previous round) - I once had to float up five rounds in a row under US pairing rules (which ignore floats), and hated it - but with the DOP having the discretion to ignore it (and announcing so at the start) if they feel more sensible pairings would be generated without it...

Bill Gletsos
18-10-2004, 12:01 PM
Charles Z here from Lee's account. A few points:

(a) I was not the one who told Lee the Round 6 pairings were wrong (I left the weekender on the Sunday at lunchtime, so was not aware of the problem until after the tournament), but agree with Bill (it happens occasionally!) that the pairings look like they were actually technically correct.
Even though from my discussion with Lee (I hadnt explicitly asked him) I was fairly certain you were not the one who claimed the draw was wrong I thought I should wait for Lee (or you) to state it.


(b) Pairings should be provisional for weekenders, but made public (and not changed except under extreme circumstances) for once a week events like the NSW State Champs so players can prepare.
I pretty much agree with this.


(c) Computers are a TOOL to aid the DOP, not the be-all and end-all, so pairings should be changed if they're wrong (as I feel Swiss Perfect often is).
I certainly agree with this (Actually CZ you and I generally attend to agree on most things!).


(d) I would change the float rules by omitting B2(d) (cannot float in the same direction as two rounds before), but keep B2(c) (cannot float in the same direction as the previous round) - I once had to float up five rounds in a row under US pairing rules (which ignore floats), and hated it - but with the DOP having the discretion to ignore it (and announcing so at the start) if they feel more sensible pairings would be generated without it...
Just to claify something here for others.
Charles reference to B2(d) and B2(c) refer to Stewart Reubens renumbering of the B rules in his 2nd Edition Chess Organisers Handbook. This renumbering was in no way official but as I explained above an attempt by Stewart to overcome some ambiguity in the C section of the rules.
B2(c) is B5 and B2(d) is B6 of the official rules as listed on the FIDE web site.

Thunderspirit
18-10-2004, 12:36 PM
Charles Z again. Thanks Bill for the clarifications. And yes, we do tend to agree on most things. But not all. Life would be dull if we did...

Bill Gletsos
18-10-2004, 12:42 PM
Charles Z again. Thanks Bill for the clarifications. And yes, we do tend to agree on most things. But not all. Life would be dull if we did...
It would indeed. ;)

Denis_Jessop
18-10-2004, 08:48 PM
Just to claify something here for others.
Charles reference to B2(d) and B2(c) refer to Stewart Reubens renumbering of the B rules in his 2nd Edition Chess Organisers Handbook. This renumbering was in no way official but as I explained above an attempt by Stewart to overcome some ambiguity in the C section of the rules.
B2(c) is B5 and B2(d) is B6 of the official rules as listed on the FIDE web site.

Hi Bill
Could you, or anyone else, please clarify something more for me. I see that SR has not only renumbered the clauses in B (which I knew he had done but thought, without checking, that it was FIDE's work) but his B1(c) re half-point byes seems not to be in the FIDE Handbook version nor is his B3 re players of the same federation. On the other hand, in Part F he has left out the words "unless the rules of the tournament state otherwise" before F8 and has left out F10 re final standings both of which are still in the Handbook. Both of these were in his first (orange) edition. As a matter of interest, I note that F10 is inconsistent with the default settings on SP.

Denis J

Kevin Bonham
18-10-2004, 09:49 PM
Ian here, now.

I don't think Kevin's precondition for dropping the B6 rule - that the tournament be of a short length - relevant. The problems in the Australian Open I mentioned earlier happened in the latter rounds of an 11 round event with a relatively modest number of participants.

I wonder how long ago that Open was and whether it happened before the following note was included:

Note: B2,B5 and B6 do not apply when pairing players with a score of over 50% in the final round.

That has been there since at least 1992 and would at least have solved the problem mentioned for the final round.

For rounds before the previous I would really need to know what year it was, and see a cross-table to comment. Under normal circumstances anyone who is on the same score as a "strong player" eight or nine rounds in in an Australian Open can't be too weak (unless the "strong player" is having a shocker.) Also, sometimes if one of the leaders gets an easy match it is because they have played a disproportionate number of other leaders along the way.

Genuine irregularities can happen whichever way you go. I'm curious about how common genuine irregularities caused by B6 (outside the kinds of tournaments I specified above) actually are, to weigh these up against what you get if you abolish the rule (namely the same player can keep getting floated in the same direction every second round, probably with the same colour.)

One thing I have found odd about B6 is that a player who has already had two downfloats widely separated cannot have two upfloats in three rounds even if that would balance out their floats for the event. However a player could be floated in the same direction every three rounds. So one possible alternative would be to have a concept of float balance similar to the colour balance concept.

Trent Parker
19-10-2004, 12:15 AM
Hey Whilst this discussion is on Swiss Pairings. Does anyone know where one might find a swiss pairing program?...... Preferably free?

Garvinator
19-10-2004, 12:19 AM
Hey Whilst this discussion is on Swiss Pairings. Does anyone know where one might find a swiss pairing program?...... Preferably free?
www.swissperfect.com

Bill Gletsos
19-10-2004, 12:22 AM
www.swissperfect.com
But not necessarily free.

Garvinator
19-10-2004, 12:24 AM
But not necessarily free.
it does have a evaluation period, then you have to pay for a registration key after that time.

Trent Parker
19-10-2004, 12:37 AM
Thanks :confused:

Bill Gletsos
19-10-2004, 11:36 PM
Hi Bill
Could you, or anyone else, please clarify something more for me. I see that SR has not only renumbered the clauses in B (which I knew he had done but thought, without checking, that it was FIDE's work) but his B1(c) re half-point byes seems not to be in the FIDE Handbook version nor is his B3 re players of the same federation. On the other hand, in Part F he has left out the words "unless the rules of the tournament state otherwise" before F8 and has left out F10 re final standings both of which are still in the Handbook. Both of these were in his first (orange) edition. As a matter of interest, I note that F10 is inconsistent with the default settings on SP.

Denis J
Yes, I had noted the differences as well.
Except for his B3 which he notes was approved at an Executive Board meeting of 1999, my attitude is that the rules as listed on the FIDE web site should the ones used.
The new FIDE 2005 handbook should be ratified by the end of the 75th FIDE Congress at the Olympiad, so hopeful they will post the latest version to the website shortly thereafter.

Garvinator
22-03-2005, 10:44 PM
Hello Bill and others,

I dont seem to have the sp files for this tournament anymore. Can someone please send me the sp files or give me a link so I can see the tournament table. Sp files would be preferred.

Bill Gletsos
22-03-2005, 11:02 PM
Hello Bill and others,

I dont seem to have the sp files for this tournament anymore. Can someone please send me the sp files or give me a link so I can see the tournament table. Sp files would be preferred.Done.

Garvinator
22-03-2005, 11:17 PM
Done.
thank you