Dramatic changes in congress performances

Discuss anything you like about chess related matters in this forum.
Brian Towers
Posts: 1255
Joined: Tue Nov 18, 2014 7:23 pm

Re: Dramatic changes in congress performances

Post by Brian Towers » Sun Nov 01, 2015 12:47 pm

Stewart Reuben wrote:Since Acceleration has never been computerised
Oh, yes it has!
Swiss Master has a very sophisticated acceleration option in terms of groups, points, number of rounds to apply, etc.
Vega (recently reviewed by Gerry Jepps) also has it. Since it is a trivial programming exercise I'd be surprised if any of the approved programs didn't have it as an option.

Since
FIDE Handbook C.04.2 General handling rules for Swiss Tournaments wrote: Accelerated methods are acceptable if they were announced in advance by the organizer and are not biased in favour of any player.
there is no reason for tournament organizers not to use it if they so wish and feel there is a need.
shaunpress wrote: I can still report that FIDE SPP is looking at developing a standardised system of Acceleration. However at this point they are still running up against the problems of how long to accelerate for
Why is it a problem for FIDE? Why is it not a problem for organizers to decide?

Of course FIDE can give guidelines. The standard (not FIDE) guidelines are, I think, that acceleration should in all cases stop at least 2 rounds before the end. Otherwise it should continue until nobody in the bottom half is on 100%.
If the aim is to reduce the yo-yoing in the first few rounds then just have it for 2 or 3 rounds (in a 9 rounder).

How popular is this, in any case, with the chess playing public? I can imagine it being popular with the minority of very strong players, but speaking for the semi-patzer level, the first round of large rapids and opens is the one time I get to have a pop at a GM or IM with perhaps the chance of some free analysis and advice (if I can actually understand and follow what he's saying) in the post mortem!
shaunpress wrote:and what to do when it stops.
Perhaps I'm being a bit thick (wouldn't be the first time) but I don't understand. Could you explain why there is more to it than just removing the "logical point(s)" used in the acceleration process and continuing with normal pairing?
Ah, but I was so much older then. I'm younger than that now.

shaunpress
Posts: 76
Joined: Mon Oct 18, 2010 3:41 am

Re: Dramatic changes in congress performances

Post by shaunpress » Sun Nov 01, 2015 12:55 pm

Roger de Coverly wrote: The French managed to use existing software and its pairing rules by use of a dummy point system. That's not to say it doesn't create all the problems outlined, to which could be added that acceleration can also give a lower rated player a short cut to being in contention for prize money, whilst facing a lower scoring field than other contenders.
The 'dummy point' system has been tried but found wanting under most circumstances (assuming you mean give the top half an extra point before round 1, but then remove it at some later stage). Removing it too early just moves the mismatches to that round (cf 2008 Olympiad round 3), while leaving it for too long distorts the tournament pairings unnecessarily. The recommended application is to keep it until no one in the lower half has a perfect score. Some people do like it though, mainly due to its simplicity.

Mike Gunn
Posts: 775
Joined: Wed Apr 11, 2007 4:45 pm

Re: Dramatic changes in congress performances

Post by Mike Gunn » Sun Nov 01, 2015 12:59 pm

Brian Towers wrote: Perhaps I'm being a bit thick (wouldn't be the first time) but I don't understand. Could you explain why there is more to it than just removing the "logical point(s)" used in the acceleration process and continuing with normal pairing?
The normal method of doing acceleration (see Stewart's book) is not based on the "logical points" approach. However, you are perfectly correct that stopping acceleration should not be a problem, regardless of how it was done.

shaunpress
Posts: 76
Joined: Mon Oct 18, 2010 3:41 am

Re: Dramatic changes in congress performances

Post by shaunpress » Sun Nov 01, 2015 1:03 pm

Brian Towers wrote:
shaunpress wrote: I can still report that FIDE SPP is looking at developing a standardised system of Acceleration. However at this point they are still running up against the problems of how long to accelerate for
Why is it a problem for FIDE? Why is it not a problem for organizers to decide?

Of course FIDE can give guidelines. The standard (not FIDE) guidelines are, I think, that acceleration should in all cases stop at least 2 rounds before the end. Otherwise it should continue until nobody in the bottom half is on 100%.
If the aim is to reduce the yo-yoing in the first few rounds then just have it for 2 or 3 rounds (in a 9 rounder).

How popular is this, in any case, with the chess playing public? I can imagine it being popular with the minority of very strong players, but speaking for the semi-patzer level, the first round of large rapids and opens is the one time I get to have a pop at a GM or IM with perhaps the chance of some free analysis and advice (if I can actually understand and follow what he's saying) in the post mortem!
shaunpress wrote:and what to do when it stops.
Perhaps I'm being a bit thick (wouldn't be the first time) but I don't understand. Could you explain why there is more to it than just removing the "logical point(s)" used in the acceleration process and continuing with normal pairing?
Essentially the issue is one of standardisation. There are now standardised pairing rules (Dutch, Dubov etc) but no "standard" way of doing accelerated pairings. As FIDE approve software to do tournament pairings (and non approved systems invalidate norms), developers need to know what system they should be implementing.

In my experience unnecessary acceleration (for non norm tournaments, or tournaments with enough rounds) is generally unpopular below the top level, for reasons you give (chance to play an IM/GM), although this in part is also due to unsatisfactory acceleration methods (two easy games and then two hard games is just a different yo-yo). This often happens when the bonus points get removed.

Reg Clucas
Posts: 402
Joined: Mon May 16, 2011 3:45 pm

Re: Dramatic changes in congress performances

Post by Reg Clucas » Sun Nov 01, 2015 1:06 pm

Roger de Coverly wrote:
Most Congress entry forms reserve the right to put players in a different section to one they've previously won.
I haven't seen any such condition on any congresses I've entered in the last two years, though admittedly this is a small sample. I remember that it used to be fairly standard for entry forms to contain a clause such as "the organisers reserve the right to move players to a different section" (i.e. anyone, not just previous winners) but I haven't seen this for a while either.

Stewart Reuben
Posts: 4186
Joined: Tue Apr 03, 2007 11:04 pm
Location: writer

Re: Dramatic changes in congress performances

Post by Stewart Reuben » Sun Nov 01, 2015 1:25 pm

Indeed, Shaun it is good to hear from you.
The French Accelerated Paring system has been available computerised for some years. It is based on the original Phil Hayley (CAN) bonus point system.
This was explained to me verbally in Atlantic City in 1964. I then used an Accelerated Pairing System for the first time at the Islington Open of 1968 when there were 170 entries. However, I either misundestood or forgot Phil's system and invented my own fractional pairing system. Christian Krause didn't like systems to be named after people, that is why the two names.
Naturally I believe my system to be superior. Certainly it is more efficient at cutting down the number of weaker players with 100%. It uses the higher rated players who draw to bash the weaker. 1968 I misfiled Les Blackstock, a 200 strength player, in the bottom group. The system still coped. Only once did a player from a lower group reach 3/3 whle I ran these events. In 1971 Peter Morrey, about 180, was lone leader on the 3/3, ahead of Kavalek and Timman (then an IM). He finished with the excellent score of 4/6.

Ricco tells me a system can be computerised with about one month's work. Then you would simulate running the same tournament perhaps 1000 times, using the actul result from a tournament that had happened. Then comparing the effect of different pairing systems.
Not much point in doing that, until you have decided your objectives.
The Kazic book of 1980, The Chess Competitor's Handbook', gives The London Chess Association system. That was prepared by the Late Peter Morrish, David Eustace, others and me.
1. Players in Swiss System events should enjoy themselves as much as reasonably possible.
2. Players generall prefer to play, rather than win by default. {Internationally that is quite controversial}
3. A player's final score should reflect his performance, e.g. similar scores should have similar sums of opponent's scores and similar rating results. Higher than that of players on lower scores.
4. A player should be able to pre-determine his opponents. 2 arbiters should independently come up with the same pairings. {That was before computers.)
5. Clashes between players of hugely disaparate strength should be minimised.
6. There is no doubt it diminishes the bouncing effect.
7. It may help players meet stronger opponents, provided they start well. This makes it more efficient if norms is a prime objective.

Have no doubt, Accelerating for only two rounds will always lead to nonsense. I was told that in 1964. I warned what was going to happen in 2008 in Dresden. I have never seen any need to continue for more than three rounds.
Strong players don't like to find their rivals meeting low rated opponents in later rounds.
Weaker players don't like it that they will only get a crack at the top players if they have performed well.

Acceleration is more complex to explain. But, trying understanding VIRTUAL OPPONENT!

These matters were to have been discussed in Abu Dhabi, but there was no time. Possibly they didn't want to discuss Fractional Pairings in my presence.

The perfectly sound reason FIDE do not want to leave it up to the organisers without computerisation is that they don't trust organisers not to fiddle the pairings. In Spain they use a system where there are FOUR different tiebreak systems. After play has finished, one envelope is chosen by lot from the four. I have been told that drawing of lots is not always done in the presence of the players and strangely the system chosen favours the local players - according to some people.

harrylamb
Posts: 147
Joined: Thu Apr 22, 2010 1:33 am

Re: Dramatic changes in congress performances

Post by harrylamb » Sun Nov 01, 2015 2:49 pm

Mick Norris wrote:
Michael Farthing wrote:
Mick Norris wrote: This has been tried, and has failed, so congress organisers who want to at least break even stick with a model that works
When? [Serious question, Mick]. [I'm thinking specifically of lower sections].
Manchester congress did this a few years back
Alasdair McLeod wrote: Just a sample of one tournament then.
No. It was at least three.

I was MCF congress director at the time. It was about 10 years ago. I put in the innovation of everyone playing in the same section. It was not a success. I tried it because I play around three tournaments a year in France. This is the French formula and I thought it would work in England. My main driver was to reduce costs. With one big section you can save on the prize fund as there are no Major and Minor main prizes. With the money I saved I put in generous rating prizes and kept entry fees unchanged in spite of other costs increasing.

I ran at least three congresses in succession like that. The outcome was disappointing. The first one people actually gave it a try although entries were down slightly. But they could have been many reasons for that. Then things got worse. By the time I held my third congress entries were considerably down. The money in the congress fund was virtually zero and I had a financial crisis. We used to run three congresses a year (Wigan Winter) (Manchester Summer) (Manchester at a different venue Autumn). We finished up running one (Manchester Summer)

When I asked my former competitors why they were not entering. The ones in the old minor, where my biggest loss of entrants was, told me that they did not enjoy the thought of being paired against the strong players. In vain I explained that the draw was accelerated and especially in the first round they were being paired against people with similar ratings as they would play against in conventional tournament. The “minor” players did not like the event and they voted with their feet.
As a result of our congress reserves dropping to zero, we scrapped the Winter and Autumn congresses and now run just one event with the traditional three sections. The upside is that I now believe I know what is needed for a successful congress. This may be controversial. But I believe.

Entry fees
Chess players are not the tight fisted skinflints we portray ourselves as being (eg on this forum.) We are prepared to pay good entry fees provided we get good facilities.

Playing Room
A good playing room with adequate space and good tables.

Players spending the weekend away
a) Good restaurants
You can see this on this forum. The moment the 4NCL announce a new venue its regular players are on this forum asking if anyone knows a good restaurant nearby
b) Good accommodation
Either on site or modern chain hotel type accommodation eg Travel Lodge, Ibis, Premier lnn, Campanile.
Before anyone berates me over the fact that the Manchester Congress on site student accommodation is rather basic. I do get 10% of the entry staying there and all the “chains” mentioned have hotels a direct bus ride away in Central Manchester and attract bookings from our entrants
c) Local tourist attractions

Not FIDE rated
FIDE rating of five round weekend tournament I believe is detrimental to entries. On the whole adult entrants do not want to put their FIDE rating at risk in the hurly burly of a weekend tournament.

I have rebuilt the Congress bank balances following these principles. Our entries have gone up from a low of 60 in 2004 to a record high of 120 in 2015. Noteworthy is

• All our equipment is new – Digital clocks-Boards-sets. And I thank Mick Norris for facilitating these purchases
• I emphasise Curry Mile and Chinatown restaurants.
• Manchester believe it or not is a fast growing tourist town. From traditional attractions like Coronation street studios to more recent ones like the Gay Village entertainment area
• I have entrants who take a half point bye on Sunday mornings because of Saturday night attractions.

Finally I believe that by accident or design all the very successful Northern congresses are following the same path. They are

Blackpool
Scarborough
The Lakes

And Bolton got a big rise in entries this year. Due I believe to a good playing venue and a new Travel Lodge built about 300 yards away
No taxation without representation

User avatar
Michael Farthing
Posts: 1961
Joined: Fri Apr 04, 2014 1:28 pm
Location: Morecambe, Europe

Re: Dramatic changes in congress performances

Post by Michael Farthing » Sun Nov 01, 2015 4:11 pm

A very interesting post Harry. Thank you.

Earlier in the thread there was a specific query about prize levels, which you have not commented on. Anything to add?

Stewart Reuben
Posts: 4186
Joined: Tue Apr 03, 2007 11:04 pm
Location: writer

Re: Dramatic changes in congress performances

Post by Stewart Reuben » Sun Nov 01, 2015 6:06 pm

http://www.chesschat.org/showthread.php ... irings-%29 for an interesting viewpoint on the two Acceleration Systems by Kevin Bonham of Australia.
Of course I like his viewpoint as he favours fractionl over bonus point.
Alex Holowczak has pointed me towards The McMahon Sstem which is not a Swiss. (It is questionable with Accelerated is a Swiss). It is widely used in Go. That, in itself, is interesting. Many years ago, John Brew, then of Grieveson Grant, said the first mention of the Swiss was its use in Go in the 15th century.
By the way, some may not realise that Accelerated Bonus and Fractional are paired identically in round 2.

Of course if you now went for solely one section, with grading prizes, it is unlikely to be successful. People have been used to graded sections in Swisses for over 40 years. But, if graded sections had never been introduced, who knows? Prior to 1968, grading limited Swisses were virtually unknown in Britain and prior to 1964 in the US.

Certainly some people don't like weekend tournaments being FIDE Rated. Others might enter only if they are FIDE Rated. In Ireland there are a couple of events which are neither FIDE, nor Irish, rated and are highly successful. The first division of the London League was FIDE Rated for a couple of years. Players asked for this practice to stop. I don't think anybody complained when it was discontinued.

Prize levels. In English-speaking countries we tend to have high first prizes with a steep gradient down. Thus £1000, 600, 400, 300, 200 would be normal. In Spain it would be more like 750, 600, 450, 300, 200, 100, 60, 40. They use tiebreaks, we split prize money where people get the same score. A gentle gradient must mean cheating is less prevalent.

France don't allow half point byes. If you want to miss a round, you have to take a loss bye. This is inhibiting me from entering Cannes, a 9 round tournament in 7 days. I think it was the late Richard Furness who introduced the concept of half point byes.

Roger de Coverly
Posts: 19082
Joined: Tue Apr 15, 2008 2:51 pm

Re: Dramatic changes in congress performances

Post by Roger de Coverly » Sun Nov 01, 2015 7:08 pm

Stewart Reuben wrote:Prior to 1968, grading limited Swisses were virtually unknown in Britain and prior to 1964 in the US.
Tournaments over Bank Holiday weekends had long been established even in 1968. If you played in one, you could expect to be placed in either a small all play all or a small Swiss, in both cases you would expect only to meet opposition of around your own strength.

The history of the Ilford Congress for example

http://www.essexchess.org.uk/History/Il ... esults.htm

harrylamb
Posts: 147
Joined: Thu Apr 22, 2010 1:33 am

Re: Dramatic changes in congress performances

Post by harrylamb » Sun Nov 01, 2015 9:16 pm

Michael Farthing wrote:A very interesting post Harry. Thank you.
Earlier in the thread there was a specific query about prize levels, which you have not commented on. Anything to add?
Thank you Michael for your kind comments. Here are my views and some facts on entry fees and prize levels. This is a long thread. So I hope I have correctly selected the points on which you want me to comment.

Entry fees and prizes 30-45 years ago
The 1972 Manchester Congress is advertised in the 1971-2 BCF year book as having three sections

Open First prize £100
Major First prize £50
Minor First prize £25
Total prize fund of £500
Accommodation at the Venue cost £1-50 a night.

It is also advertised as being the largest weekend prize fund in England

If you convert these to 2015 money using the RPI you get

Open First prize £1090 (£340)
Major First Prize £545 (£340)
Minor First Prize £273 (£340)
Total prize fund £5453 (£2000)
Accommodation £16 (£39)

The first figure is the RPI adjusted figure for 1972. The figure in brackets is the actual figure for the 2015 event 43 years later

But I do not think that is a fair comparison. In 1972 the Manchester Congress was the biggest weekend congress in England. Now it is a shadow of its former self. I think you should compare it with the biggest current weekend tournament which is Blackpool. (Apologies to Scarborough if I am wrong). The entry form for the 2016 Blackpool event gives the following comparison. Again the first figure is RPI corrected values for 1972 Manchester and bracketed actual 2016 Blackpool.

Open First prize £1090 (£1500)
Total prize fund £5453 (£7250)
Accommodation £16 (£72)
There are actually five sections in Blackpool 2016. So I cannot give comparable Major and Minor figures

From these figures I do not think it is true to say that

“One problem with chess is that £100 prize is the same as it was 30 years ago.”

The figures to me demonstrate that at the biggest events the total prize fund after inflation has significantly increased over the last 40 years.

Effect of Prize fund and entry fees on Entries
I used to believe that increasing prizes increased entries. But I think this is no longer true and the relationship nowadays is more complicated. I think the prize fund has to feel right for the level of entry fees.

Players (and their families) want an enjoyable weekend in pleasant surroundings with an emphasis that is not 100% on chess. They are happy to pay increased entry fees to achieve this.

Cheaper entry fees and smaller prizemoney in the lower sections

This is a quote from one of Alasdair’s posts. My experience is that cheaper entries and smaller prize money in lower sections does not work. Why should it work? What is so different about the minor that should make the finances different? The blunt answer is nothing. Alasdair seems to be suggesting it to solve what I see as a non existent problem of the winner of the Open getting the same amount as the winner of the minor. Plus the GM he quotes who got £7.50 for coming joint third may not be a happy bunny. But reducing the minor player’s entry fee by £15 and not paying the minor player any prize money will not solve the GM’s problem.

Entry fees are there to fund the tournament. As a congress director I have 120 people at each Congress asking me to organise a weekend’s pleasure for them and that is a big responsibility. They want
• Good playing conditions.
• An enjoyable evening out on the Saturday night.
• Entry fees commensurate with the cost of the event
• A traditional prize structure.

That is what I try to give them and that is why they enter

Players want to win prizes
Players want to win a prize. The value of the prize is not important. They want to be able to go home and proudly say I spent the weekend playing chess. I won a prize!
No taxation without representation

Ian Thompson
Posts: 2584
Joined: Wed Jul 02, 2008 4:31 pm
Location: Awbridge, Hampshire

Re: Dramatic changes in congress performances

Post by Ian Thompson » Sun Nov 01, 2015 10:56 pm

harrylamb wrote:But I do not think that is a fair comparison. In 1972 the Manchester Congress was the biggest weekend congress in England. Now it is a shadow of its former self. I think you should compare it with the biggest current weekend tournament ..

From these figures I do not think it is true to say that

“One problem with chess is that £100 prize is the same as it was 30 years ago.”

The figures to me demonstrate that at the biggest events the total prize fund after inflation has significantly increased over the last 40 years.
It would be interesting to know what the comparison is for other events that have been running for 30 to 40 years, e.g.:
  • British Championships
    Hastings
    Guernsey
    Paignton
    Union Championships
    Dorset
    Herfordshire
    Torbay
    Kidlington
    Nottingham
    Probably a few other weekend congresses

Stewart Reuben
Posts: 4186
Joined: Tue Apr 03, 2007 11:04 pm
Location: writer

Re: Dramatic changes in congress performances

Post by Stewart Reuben » Mon Nov 02, 2015 1:57 am

Harry Lamb > In 1972 the Manchester Congress was the biggest weekend congress in England.
There must be a failure in English here. Perhaps what was meant was the biggest so far in 1972 in England. The Evening Standard London Chess Congress had 1200 entries. But it took place in December.
In 1978 the National Bank of Dubai Open in the Evening Standard Congress by then in July and in later years also the Aaronson Brothers Open in Harrow in September both had a first prize of £1200. I remember Bill Hartston commenting that the first prize for a weekend tournament should not be higher than that for the British Championship.
I suppose I was referring to the rapidplay tournaments many of which still have a £100 first prize.
It is a very long time since I organised a weekend tournament. But, at the time I did so, it seemed that, if the entry fee was increased, the number of entrants increased.
Also I have a London-based perception. There is no doubt there is less activity here today than 40 years ago, with the exception of the wonderful London Chess Classic and Adam Raoof's events.
Until the Sims Prize of about £300 came along, the first prize in the British was £35. The nature of Hastings has changed as has that of the British. In both those events top players receive start money. In 1997 the Smith & Williamson British Championship had a first prize of £10,000 - but no start money. Now the first prize is £5000, but there is start money for GMs.
The Gibraltar prize money is inflation busting of course - and open to all. But it only started in 2003.

User avatar
Jon Mahony
Posts: 606
Joined: Sun Jun 14, 2009 10:47 pm
Location: Leeds
Contact:

Re: Dramatic changes in congress performances

Post by Jon Mahony » Mon Nov 02, 2015 1:01 pm

Reg Clucas wrote:
Roger de Coverly wrote:
Most Congress entry forms reserve the right to put players in a different section to one they've previously won.
I haven't seen any such condition on any congresses I've entered in the last two years, though admittedly this is a small sample. I remember that it used to be fairly standard for entry forms to contain a clause such as "the organisers reserve the right to move players to a different section" (i.e. anyone, not just previous winners) but I haven't seen this for a while either.
Blackpool does tend to do this still, but I know what you mean, I've been allowed back in several Minors and Inters I've won in the past. Scarborough used to be quite strict on this too but that seems to have been relaxed - Dave Dunn and of course Mr Crockett have been allowed to play the minor year after year. Also my friend Dave Summerland has won it 5/5 and he has been allowed in the section below even - though this is probably fair, he has been out of form for a good few years now and his grading has gone as low as 107 recently.

I was asked to move up to the major in York in the last year it ran but I found it outrageous as the organisers reason for doing it was the fact I’d recently beaten a few 140’s and my live YCA grade (which means nothing till the end of the year) had gone a whole point above the limit - ECF grade was 5 points below the limit.

Think they could see a full 5 pointer (it was points mean prizes) and they’d actually have to pay some money :lol:
"When you see a good move, look for a better one!" - Lasker

Paul Bielby
Posts: 152
Joined: Tue Dec 22, 2009 4:14 pm
Location: South Shields

Re: Dramatic changes in congress performances

Post by Paul Bielby » Mon Nov 02, 2015 1:12 pm

My first congress was the Yorkshire Easter Congress, played in Huddersfield in 1953. Organised by Edgar Priestley, it had 60 entries, divided into 10 all-play-all sections each of 6 players. This was, of course, before the days of gradings so the organiser had to arrange the sections according to the best knowledge of the players he had. I was in the bottom section for Juniors. The entry fee was £1. The first prize was £3 and second £!/10 shillings. I won three games, drew 2 and finished 1st= with two other boys. We were pleased to win £1.50 each! The winner of the top section was T.K.Hemingway, with the then British Champion, Bob Wade second. I don't know if their prize money was different from ours, but it can't have been much.

The Yorkshire Easter Congress continued in the same fashion at least until about 1970 - it was the first tournament I ever organised, when it returned to Huddersfield in 1967 and we still used 6 player all-play-alls then (I did have gradings to help me by then). At some point in the '70s they changed to Swiss tournaments.

It is probably due to its evolution from many little all-play-alls where all players had much the same grading, that Swiss tournaments started with graded sections.

Since retirement I have recently played in large one-section Swiss tournaments abroad. One can't help noticing how quickly their juniors, thrown in at the deep end into open tournaments improve, as compared with ours in this country. This may account for the modest performance of our youngsters abroad such as the current WYCC in Greece. I am afraid that our system, with big prizes in minor tournaments, simply encourages mediocrity. But that's probably a topic for another thread.

Locked