Page 1 of 3

White wins in two moves and other matters

Posted: Wed Dec 21, 2016 8:00 pm
by DonaldMoir
I am only a very 'average' chess player, so I hope you don't mind if I share some recent chess experiences with you all.

I had the enjoyable task of hosting a group of ten Indian children (aged 10 - 15 years) and two of their teachers, when they attended the London Chess Classic last week. [As well as a week-end tournament and the Super RapidPlay tournament, they all played John Nunn in his simultaneous display, so you can see them in the excellent photographs on the LCC website.] The children all thoroughly enjoyed themselves and remarked on the efficiency and friendliness of the event.

One of the boys, Vijaya Siva, aged 10 years, obtained, I imagine, the quickest victory of the whole event. He was white. 1. c4, e5; 2. g3, Nb6?? 1-0

Being retired and being friends with their school Principal, I have visited Keren School (near Madurai) in Tamil Nadu on several occasions over the past five years and have helped to organize their chess. Two periods/week are assigned to electives, including chess, which is chosen by about 200 students. Ebenezer Joseph from Chennai comes once a month to give overall guidance. Twice they have performed human chess matches, and Deepan Chakhravarthy and RR Laxman (+ two others) gave simultaneous, simultaneous displays (4 x 20 = 80 students) and, most enjoyably, Vishy Anand came to the school just after last years LCC. Regularly, up to 100 students attend tournaments in Madurai and elsewhere in Tamil Nadu.

My comments and questions are:
1. In England, compared with India, there seems a remarkable dearth of players with FIDE ratings between 1000 and 14000. In FIDE-rated tournaments in India there are, literally, hundreds of players who have these grades.
2. In my experience, (admittedly subjective) there seems at least a 200 gap between Indian and English grades - i.e. an 'Indian' FIDE rating of 1200 seems to be about equivalent to an 'English' grade of 1400-1500. Most of the Indian boys gained useful increases in their ratings. Presumably this is because, at lower grade levels, there is almost no mixing between the two chess communities.
3. One small problem, but which causes distress to a few individual beginners, is that - both in India and England - the selection of a person to be given a bye when there are an odd number of competitors is not done randomly but always by picking the player at the end of the alphabet. Last week one of our children, named Vicksho, was sad to miss the second game out of his five for this reason. In India, a child called Yojith, was so regularly selected for the bye, that he stopped coming to tournaments.
4. In India, and presumably England, there are not that many FIDE-rated rapid tournaments. Thus many very experienced players have 'falsely' low rapid grades. Would it not be better to set the grading prize level, so that eligible players need to be below the bar in both rapid and standard grades? In last week-end's tournaments, us genuinely lowly graded individuals really had no chance of competing for a grading prize.

But these are small issues. Well done to Malcolm Pein and all those who organized such a splendid event.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:09 pm
by Andrew Zigmond
DonaldMoir wrote:

My comments and questions are:
1. In England, compared with India, there seems a remarkable dearth of players with FIDE ratings between 1000 and 14000. In FIDE-rated tournaments in India there are, literally, hundreds of players who have these grades.
2. In my experience, (admittedly subjective) there seems at least a 200 gap between Indian and English grades - i.e. an 'Indian' FIDE rating of 1200 seems to be about equivalent to an 'English' grade of 1400-1500. Most of the Indian boys gained useful increases in their ratings. Presumably this is because, at lower grade levels, there is almost no mixing between the two chess communities.
3. One small problem, but which causes distress to a few individual beginners, is that - both in India and England - the selection of a person to be given a bye when there are an odd number of competitors is not done randomly but always by picking the player at the end of the alphabet. Last week one of our children, named Vicksho, was sad to miss the second game out of his five for this reason. In India, a child called Yojith, was so regularly selected for the bye, that he stopped coming to tournaments.
4. In India, and presumably England, there are not that many FIDE-rated rapid tournaments. Thus many very experienced players have 'falsely' low rapid grades. Would it not be better to set the grading prize level, so that eligible players need to be below the bar in both rapid and standard grades? In last week-end's tournaments, us genuinely lowly graded individuals really had no chance of competing for a grading prize.

But these are small issues. Well done to Malcolm Pein and all those who organized such a splendid event.
A few comments on points 1-3.

1, I think this is simply because there is no call in this country for FIDE rated chess to be played at this level. There are plenty of opportunities for lower rated players to play chess in the UK but such events use the national grading system. To be fair, FIDE rated tournaments for all levels are relatively few and far between (and E2-E4 the only major tournament provider that insists upon FIDE rating for all levels) and FIDE rated chess necessitates a higher level of ECF membership.

2, A statistician will probably be able to give a better answer but a grade measures your performance against other players. It does not give a definitive indication of playing strength.

3, Is this the case in England? I only remember getting the bye once as a junior (in the fifth round of an event) and have only had it once as an adult (in the fourth round of a congress this year) and both times I was dead last. Obviously I tend to be the last competitor alphabetically. To be fair I turned 18 last century so things may have changed. (EDIT following Roger's post below my junior days pre date the pairing rules and obviously my position in any pairing structure depends on my rating, rather than my surname).

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:10 pm
by Roger de Coverly
DonaldMoir wrote: 1. In England, compared with India, there seems a remarkable dearth of players with FIDE ratings between 1000 and 1400
There are plenty of players, they just don't have International ratings. The ECF has a long established grading system of its own, so every club and tournament player has a grade.
DonaldMoir wrote: 3. One small problem, but which causes distress to a few individual beginners, is that - both in India and England - the selection of a person to be given a bye when there are an odd number of competitors is not done randomly but always by picking the player at the end of the alphabet.
You can blame FIDE for that one. Left to their own devices, running tournaments to their self invented rules rather than FIDE's, British organisers would either avoid byes entirely by making a substitute available or by awarding the bye to the middle rather than the bottom of a score group. Also they would rank players by organiser's estimate so unrated players were not ranked solely by alphabet.

Even under FIDE rules, there's a trick for floating the median if allowed. You insert a player Zzzz in the tournament with a rating of zero. That makes the numbers even, thus the last board features a median player against Zzzz.
DonaldMoir wrote: 4. In India, and presumably England, there are not that many FIDE-rated rapid tournaments.
In England, there are numerous rapid tournaments, graded domestically. Not everyone plays in them with any regularity.

There was a hierachy to the ratings used in the Super Rapidplay. I believe it went something like
FIDE rapid
ECF rapid (converted)
FIDE standard
ECF standard (converted)

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:16 pm
by Roger de Coverly
Andrew Zigmond wrote: 3, Is this the case in England? I only remember getting the bye once as a junior (in the fifth round of an event) and have only had it once as an adult (in the fourth round of a congress this year) and both times I was dead last. Obviously I tend to be the last competitor alphabetically.
The FIDE rules on pairing applied strictly will rank players by rating and then alphabetically within rating. This has the side effect of treating all players without ratings as having ratings of zero and then ranking them alphabetically. That's the way it's done at these big international Junior events. It's commented sometimes in reports on these events that it's better for players to avoid this by already having an established rating.

For adult events, it's much more likely that players without International ratings will be given estimates for pairing purposes based on their national grades or ratings. Unlike England (or Yorkshire), India doesn't have a domestic rating or grading system.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:26 pm
by NickFaulks
Roger de Coverly wrote: You can blame FIDE for that one.
Even by the standards of your fevered imagination, that is a remarkable statement. What on earth do you believe FIDE regulations have to say about awarding byes based on alphabetical order?

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:28 pm
by Ian Thompson
DonaldMoir wrote:Last week one of our children, named Vicksho, was sad to miss the second game out of his five for this reason.
I see he got a bye in Round 2 of the Weekend U1600 event, but played all his other games, so I'm not sure which event you're referring to.

In any case, the FIDE pairing rules prohibit a player from getting a bye if they've had a free point, or half point, already, for any reason. For example, you can't be given the bye twice in a tournament, nor can you be given the bye if you've previously had a walkover win, or taken a half-point bye.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:40 pm
by Roger de Coverly
NickFaulks wrote: What on earth do you believe FIDE regulations have to say about awarding byes based on alphabetical order?
Don't you realise why?

FIDE pairing systems work by ranking the players. The effect of this is that the bottom ranked player will get the bye if there's an odd number. The CAA British system avoids this by floating or giving the bye to the median.

Next ask yourself who is the lowest ranked player? Obvious if every player has a rating, less so if they don't. Particularly in large Junior events it's the established practice that players without ratings are treated as having a zero rating and are ranked in alphabetic order. Now think of the likely consequence of being both unrated and at the end of the alphabet. Do you not think such a player would be most likely to get the bye and very possibly in repeated events until they escaped by getting a rating? Now think about how to avoid this? How else would you do so without changing the FIDE pairing rules?

So it is the fault of those who devise FIDE regulations and arguably arbiters who feel obliged to apply them in an inflexible manner. The problem of those at the end of the alphabet being favoured or plagued with byes would be alleviated by assigning pairing numbers for the unrated in a random manner to follow the rated. The pairing programs will do the rest.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 9:50 pm
by DonaldMoir
Yes, Vicksho played in the U1600 W/E tournament and, because of the bye, missed one of his five games. Fortunately, he got a win in time to avoid a bye in the Super Rapidplay.

It's not a big deal, but it happens to him, Yojith, and a few other kids unfortunate enough to have been given low-alphabet names on a regular basis.

I am not sure why the Swiss pairing program cannot simply chose the bye player randomly from those on the lowest points level. The advice to avoid this by getting an official grade is easier said than done. Vicksho has only been playing chess for one year.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 10:33 pm
by NickFaulks
Roger de Coverly wrote: Don't you realise why?
We have had similar discussions before. Just because FIDE cannot think of every possible discriminatory way of producing pairings while staying within the letter of their regulations and ban them individually, that does not mean it is their fault when organisers go out of their way to do that.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 11:14 pm
by Roger de Coverly
NickFaulks wrote: that does not mean it is their fault when organisers go out of their way to do that.
The ethos of FIDE's monitoring of pairing methods is that it's very much opposed to arbiters using tricks to amend the pairings in some manner or other. Given FIDE's continued enthusiasm for awarding titles as prizes, why is it so opposed to arbiters doing a little rule bending on occasion?

The journal of the CAA is suggesting a potential crisis for their pairing method. Unless they can prove it is 100% deterministic by getting a program written to implement it, it's liable to be outlawed, for Norm tournaments at the very least.

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 11:17 pm
by Graham Borrowdale
DonaldMoir wrote: One of the boys, Vijaya Siva, aged 10 years, obtained, I imagine, the quickest victory of the whole event. He was white. 1. c4, e5; 2. g3, Nb6?? 1-0.
That was a bit of a shame, and I am guessing a disappointment to the 10 year old boy - nobody wants to win a game like that. I presume it was a default for making an illegal move?

Re: White wins in two moves and other matters

Posted: Wed Dec 21, 2016 11:18 pm
by Roger de Coverly
DonaldMoir wrote: I am not sure why the Swiss pairing program cannot simply chose the bye player randomly from those on the lowest points level.
Technically that's obviously not difficult, but FIDE rules preclude it.

It may be worth asking the organisers of the Weekend tournaments why they felt it necessary in a tournament for less experienced players to apply rules that discriminate against unrated players with names at the end of the alphabet.

Re: White wins in two moves and other matters

Posted: Thu Dec 22, 2016 12:51 am
by Andrew Zigmond
Roger de Coverly wrote:
DonaldMoir wrote: I am not sure why the Swiss pairing program cannot simply chose the bye player randomly from those on the lowest points level.
Technically that's obviously not difficult, but FIDE rules preclude it.

It may be worth asking the organisers of the Weekend tournaments why they felt it necessary in a tournament for less experienced players to apply rules that discriminate against unrated players with names at the end of the alphabet.
Arbiters have increasingly moved towards computer generated pairings in recent years, indeed I recall a certain prolific contributor to this forum being quite critical of those who continued to use manual pairings! It is true that some tournaments may have lower sections that are not FIDE rated (the 4NCL Congresses are the best example although the lowest weekender section at the British was not FIDE rated) and could adjust the pairing methodology, or use the Zzzzz get out but that would possibly lead to criticisms that they were being inconsistent.

In any case I believe I'm right in saying that the alphabet problem only comes into play where a large number of players are ungraded, such as junior tournaments or a rare FIDE rated low rating tournament where many players might not have FIDE grades. Donald Moir's original point was that some junior players are being discouraged due to the rule; to me the question is how many junior tournaments are FIDE rated and/ or have to pair according to strict pairing regulations?

Re: White wins in two moves and other matters

Posted: Thu Dec 22, 2016 6:29 am
by Roger de Coverly
Andrew Zigmond wrote:Donald Moir's original point was that some junior players are being discouraged due to the rule; to me the question is how many junior tournaments are FIDE rated and/ or have to pair according to strict pairing regulations?
On International ones, it applies to all of them and it would seem that the rule is also being applied to a FIDE rated Minor. The Indian experience is that it applies regardless. I'd suspect it would be less common in the UK as some arbiters would use CAA rules or just make it up as they went along. The CAA "float the median" rule would award byes to the middle of the alphabet if they used that to set rankings.

Nick Faulks seems to think the problem isn't the FIDE rule, but the insistence of arbiters in following the alphabet when setting the ranking order for players without ratings. That's where the problem lies, once you set the initial rankings, the rest of the pairings fall into place. It's something Stewart Reuben notes, that if you want more randomness in pairings, just scramble the initial ranking order.

Although not all arbiters would agree, it's better that pairings are deterministic once the initial ranking is set. Any method involving arbiter discretion whilst the tournament is in progress is open to suggestions of bias.

Re: White wins in two moves and other matters

Posted: Thu Dec 22, 2016 7:43 am
by NickFaulks
Roger de Coverly wrote: but FIDE rules preclude it.
You keep saying things like that, but when challenged you're never able to point to the actual rule which you claim exists.