Hastings Results

[Alex Holowczak]

OK, I'll bite - why does a long tail negatively impact norm chances?

Because the norm-seeker might have to play them!

To get a norm you need to score so many points. This number of points is dependent on the average rating of your opponents. So if you play a low-rated player, this average goes down, and thus you need to score more points. If you win against them, then you might end up against the GMs later on, so you still play the same sort of difficulty of opponents, but now you need to score more points than you originally did.

To get around this problem, FIDE introduced the concept of a rating floor for norm purposes. For example, if I'm a GM norm seeker, and I beat a 1890 (or an ungraded player!), then the rating gets set to 2200 for the purposes of calculating the average rating. However, you can only do this once. By having a long tail, you risk the player on 1/2 (who maybe beat a 1800, lost to a 2500) having to play someone else on 1/2. If this person is rated about 1600, then you're burdened with the 1800-rated guy counting only as 1800, not the 2200 floor.

For 10-round or 11-round events, this isn't so much of a problem, because you can discard wins against weak opponents and still make a 9-round norm, for instance. You need 27 rounds worth of norms for the title. A 9-round event doesn't offer this opportunity.

By having a section with an Open top but a bottom restricted to (say) 2000, then any potential norm seekers ought to play a stronger average field, and thus have a more likely chance of achieving his norm. You're also far more likely to get games against the correct number of titled opponents, because there will be a higher percentage of titled opponents in the field. (You'll have just kicked out a bundle of untitled players.)

[Jonathan Rogers]

This reminds me that David Coleman came 1st-2nd = in his grading section at the LCC (and gained grading points) but still lost 5 or 6 ELO rating points - because he played 3 unrated players!

[Paolo Casaschi]

London is reaching the sort of situation where a minimum rating for the "Open" tournament should be seen as almost necessary - there is no real reason why an extra section couldn't be introduced for lower rated players now that so many people are getting FIDE ratings.

I completely disagree. I like very much the "open" and "non-accelerated" nature of the London Open; while I see the advantage for norm seekers in introducing a minimum rating for entry, I believe it would be a mistake (and unfair to the rest of the amateur players) to design the whole event to meet the needs of an handful of norm seeking players.

On a side note, I always found the discussion about setting a minimum rating for entry very amusing: while everybody agrees how that threshold value should be calculated, there is always a complete disagreement about the actual threshold value, the unanimously accepted calculation rule being to "set the minimum rating for entry as one point below my own rating at the time of the tournament"

[Matt Mackenzie]

You can apply the same principle to grading/rating limited sections - save, of course, that the cut-off level should be one point higher than one's own

[Stewart Reuben]

Have you noticed that Gibraltar has had to close entries to the Open? The venue can't take more than 240 players. There is plenty of room for players under 2250 in the morning. The first chess event the Gib hoteliers ever saw was Hastings 2001-2002.

Concerning norms. It is obvious that the lower the score required, the more likely it is that people will get norms. Contenders are less likely to have to play each other. Of course norms don't have to be a main objective, but they have been very successful in attracting players to events.

Wang Yue in Hastings never got more than half a point ahead of his rivals. Would you have thought it better he won the event with a round to spare?

[IM Jack Rudd]

Yes we must look to our verb herding, and if our male herding implement is not big enough we must implement it with a french implant, wow the removablness of this horrific French import must be politically elementalized forthwith because politics has nothing to do with the male, it is purely to do with the illegality of impure artificial female breast imports.

I'm sure this is a terrific insight into the conversation. I'm just not sure that you've put it into the conversation it is a terrific insight into.

[Paolo Casaschi]

Wang Yue in Hastings never got more than half a point ahead of his rivals. Would you have thought it better he won the event with a round to spare?

As an amateur in the hopeless 2000 Elo range, when playing a tournament I could not care less about the tournament winner or about the norms scored. I care about good playing conditions and an "interesting" set of opponents. I find than an un-restricted and un-accelerated open like the London Open gives me a set of more "interesting" opponents compared to other kind of events. Just a personal choice.

As a spectator, you might be right, but I prefer the to be happy as a player rather than as a spectator.

[Stewart Reuben]

Pablo Casaschi >As an amateur in the hopeless 2000 Elo range,<
2000 is not hopeless at all. 1600 is the approximate median rating for serious chessplayers. You do yourself a disservice.

One thing I am interested in. Do you not find the bouncing effect in open seeded Swisses unpleasant? e.g. in Bermuda I beat a 1400 with white, lost to a GM with black, beat a 1600 with white, lost to an IM with black and beat an 1800 with white in the last round. Of course players at the extremes of the ratings in such tournaments don't suffer from this.

[Richard Bates]

Wang Yue in Hastings never got more than half a point ahead of his rivals. Would you have thought it better he won the event with a round to spare?

As an amateur in the hopeless 2000 Elo range, when playing a tournament I could not care less about the tournament winner or about the norms scored. I care about good playing conditions and an "interesting" set of opponents. I find than an un-restricted and un-accelerated open like the London Open gives me a set of more "interesting" opponents compared to other kind of events. Just a personal choice.

As a spectator, you might be right, but I prefer the to be happy as a player rather than as a spectator.

Don't you think that players a bit higher up the strength chain also have a right to aspire to playing an "interesting" set of opponents? Because that is what they are potentially denied in a tournament like London. An extreme, but by no means generally isolated example from London - the 17th seed (Aaron Summerscale) finished 3rd-7th and didn't play anyone above 110 points less than his own rating. Personally I think that something is not quite right for that to happen. I finished in the score group above where i was expected having played the top seed (250pts above me) and nobody else higher than 170pts below me. Partly my fault for a mid-tournament catastrophe but still doesn't seem quite right for the integrity of the tournament.

The idea of having a "crossover" rating for entry criteria is IMO a good one. There will always be people who are motivated by the prospect of winning money, being a big fish in a small pond, and others who are more motivated by the prospect of playing at least some stronger players (although I think most are satisfied with 'stronger' giving them a realistic chance of something other than lucky success - ie. they don't necessarily worry about meeting opponents many hundred of points above their rating. There is also the issue of the yo-yo effect that Stewart describes - players would rather play consistently against players around their own strength than jump between players they are expected to lose to very easily or beat very easily).

By having a crossover of, say 100pts eg. minimum rating for the top tournament of 2000, maximum rating for the lower tournament of 2100 you can provide for both whilst maintaining the attractiveness of the tournament for higher rated players, whether they are just playing for fun or are seeking norms. London has everything going for it in terms of playing conditions, location, scheduling which is why it attracts such a large entry. But the tournament itself is a bit of a relative dud.

It is interesting though how the views of a 2000 strength player differ from those a bit weaker. It is true that the 2000 strength player is from one perspective often hard done by/lacking for choice in English chess - restricted to be forever competing in Open tournaments with little prospect of ever winning prize money and having to find sole motivation in playing interesting and/or stronger opponents. One only has to follow discussions on the merits of 'favouring' the Open sections in weekend congresses to find that below about 160 ECF level opinions often become somewhat different. The idea of having one big open tournament with the main prizes going to the top players and only lower grading prizes for the weaker players is i think not very popular.

The other way in which London arguably got it wrong in principle (alluded to in another post above) was in the way it allocated rating prizes - using the traditional English method of pts scored rather than variance on rating performance W-We. Although again this has its own issues when players are regularly running into players several hundred points out of their rating category.

[Alex Holowczak]

An extreme, but by no means generally isolated example from London - the 17th seed (Aaron Summerscale) finished 3rd-7th and didn't play anyone above 110 points less than his own rating. Personally I think that something is not quite right for that to happen. I finished in the score group above where i was expected having played the top seed (250pts above me) and nobody else higher than 170pts below me. Partly my fault for a mid-tournament catastrophe but still doesn't seem quite right for the integrity of the tournament.

The thing that was not quite right may have been Aaron's own mid-tournament catastrophe. Well, I guess at his level, 4 consecutive draws against players rated 2219-2283 can be described as a catastrophe! That would have put him down the order. Still, after 6 rounds, he was tied with 3 other GMs on 4/6. Compared to those other GMs, he had the "easiest" subsequent draw of them, but I think that was more by luck than anything else. The other GMs all played 2400s.

There is also the issue of the yo-yo effect that Stewart describes - players would rather play consistently against players around their own strength than jump between players they are expected to lose to very easily or beat very easily).

There are two solutions to the yo-yo. Stewart's preferred solution is accelerated pairings. I guess the crossover is another solution, and it's certainly my preference of the two.

It is interesting though how the views of a 2000 strength player differ from those a bit weaker. It is true that the 2000 strength player is from one perspective often hard done by/lacking for choice in English chess - restricted to be forever competing in Open tournaments with little prospect of ever winning prize money and having to find sole motivation in playing interesting and/or stronger opponents. One only has to follow discussions on the merits of 'favouring' the Open sections in weekend congresses to find that below about 160 ECF level opinions often become somewhat different. The idea of having one big open tournament with the main prizes going to the top players and only lower grading prizes for the weaker players is i think not very popular.

As someone "below about 160 ECF level", and 1680 in the latest FIDE-rating list, I find my position somewhat awkward. I'm probably one of the lowest rated and weakest players who would turn up to play a FIDE-rated event, and that won't change if you make a second section a U2100 event. (Indeed, I'm often the prime candidate to be a filler, with my rating usually being perfect for someone who's the bye in the FIDE Open anyway!) It won't make much difference to me what the top limit is, because I know I'll end up playing people between 1600-1800 or so for much of the tournament. The other problem down here is that you might get unrated games, because not everyone has 9 rated games yet. I don't think this will be solved by either section limit.

There were 20 or so players for the U160 Amateur, which ran over 5 afternoons and wasn't on the official entry form (or I believe, advertised on Chess & Bridge). If you had a U2100 event, you could perhaps lose the need for the U160, and the two events could in effect merge. The U160 experiment showed that there were people willing to play for the week, but were perhaps put off by playing in the Open because they'd be seed 251 out of 260.

[Roger de Coverly]

There is also the issue of the yo-yo effect that Stewart describes - players would rather play consistently against players around their own strength than jump between players they are expected to lose to very easily or beat very easily).

There are two solutions to the yo-yo. Stewart's preferred solution is accelerated pairings. I guess the crossover is another solution, and it's certainly my preference of the two.

I don't think acceleration removes the yo-yo effect, it just changes the sequence. For a third quartile player under acceleration, the natural sequence becomes win against Q4, lose against Q1/Q2, lose against Q1/2. For rounds 4 onwards after the acceleration ends you play 100-250 points up if doing well and 100-250 down if doing badly. In an event without acceleration, the likely sequence is loss to Q2, win against Q4, loss to Q2.

The French now have a format where 7 of 9 rounds are accelerated. They use a dummy point system of acceleration to facilitate computer pairings. In some manner the dummy points are reduced or added during the event so you avoid leaders from the bottom half not meeting any other leaders. I don't know of a website that describes or explains the rules. The standard French approach is that it should be all one tournament. This is taken to extremes at Cappelle where they have over five hundred entries in the same section.

I guess the crossover is another solution, and it's certainly my preference of the two.

Go back in time at Hastings twenty years and they used to have two events. So there would be the Challengers where nearly all the field would be 2200 plus and a parallel event over an equivalent number of rounds for non FIDE-rated players, restricted to something like under 175 or 190. Those not wishing to play in the Hastings Masters and stay for the entire length of the event now play in a succession of 5 round tournaments.

From memory, it became named the Harry Baines tournament and then the World Amateur. When this moved away from Hastings, the Harry Baines was not reinstated, so the Challengers, as it then was, gained a tail.

[Paolo Casaschi]

2000 is not hopeless at all. 1600 is the approximate median rating for serious chessplayers. You do yourself a disservice.

I meant, 2000 sounds about right for me, with some luck I can aim towards 2100 maybe, but I dont have much hope to seek title norms any time soon. In other words, I'm not a young kid in transit, I'm likely to stay in this range for a while.

One thing I am interested in. Do you not find the bouncing effect in open seeded Swisses unpleasant? e.g. in Bermuda I beat a 1400 with white, lost to a GM with black, beat a 1600 with white, lost to an IM with black and beat an 1800 with white in the last round. Of course players at the extremes of the ratings in such tournaments don't suffer from this.

I'm don't mind about that. I enjoy very much playing stronger opponents and I don't mind doing my part with weaker ones. In the last two editions of the LCC I can't say I had any game ended without a good fight. For playing opponents of my same strenght I join a couple of local leagues, where my opponent is very often within 10 ECF points.

Believe me, I'd rather play only people stronger than me in an open with a threshold 1 rating point below mine, but that is too much to ask.

I entered the LCC this year with a 1996 rating, I would not have paid the entry fee and taken a week off work to play in an event "up to 2000 FIDE" or even "up to 2100 FIDE". Just count how many players below 2000 joined the Open, I suspect I would not have been the only one staying home if not allowed to play with everyone else.

However, next year it's different, since now I have 2005 Elo, so feel free to limit the Open to 2004 Elo and more

[Paolo Casaschi]

Don't you think that players a bit higher up the strength chain also have a right to aspire to playing an "interesting" set of opponents? Because that is what they are potentially denied in a tournament like London.

Sure, you pick your tournaments, I pick mine.
But don't expect the weaker players to support with their participation and their entry fees events where they are only allowed to play in the B tournament.

An extreme, but by no means generally isolated example from London - the 17th seed (Aaron Summerscale) finished 3rd-7th and didn't play anyone above 110 points less than his own rating.

Well, from round 3 to round 6 he made four draws against 22xx players, if he had scored better against them he would have had better opponents.

The idea of having a "crossover" rating for entry criteria is IMO a good one.
...
By having a crossover of, say 100pts eg. minimum rating for the top tournament of 2000, maximum rating for the lower tournament of 2100 you can provide for both whilst maintaining the attractiveness of the tournament for higher rated players, whether they are just playing for fun or are seeking norms. London has everything going for it in terms of playing conditions, location, scheduling which is why it attracts such a large entry. But the tournament itself is a bit of a relative dud.

If I had a 2400 rating I would probably agree with you.
However, look at the participation to the FIDE Open and the relatively low participation to the other minor events: could the organizers afford losing the weaker players they would lose by excluding them from the "real" Open FIDE?

Depending on the answer, you and I will make our decision about participating next year.

[Stewart Reuben]

There are many different possible scenarios.
The British Championship plus Major Open plus various 5 round events. that meets th overlap suggested.
Hastings Open plus 5 round events. The second section 9 round event was discontinued due to lack of interest. The 5 round events are sections of about 16 and thus players play their opponents within a relatively narrow range. Of course the prize money is tiny. But a 10 year old got the best rating improvement prize in the Hastings Masters and that was £200.
Gibraltar is different again. Open has gone from 2700-700 10 rounds. Two 5 round U2250 and two 5 round U1800 these four in the morning. Some U1800 play in that event plus the Open. Of course the prize money is much higher, there is even £3000 £2000 rating improvement prizes 2500-2599.
Gibraltar will remain open because then it can be boasted to be the strongest open in the world. If it were restricted then Aeroflot, would probably beat it. But there is a problem, the venue can only hold 240 players.

It was said I favour Accelerated Pairings. Certainly I do over a seeded Swiss. But Dubov and possibly Burstein are superior. I would probably favour top half v bottom half randomly in round 1, followed by seeded for the remainder over a seeded Swiss from round 1. A seeded Swiss is so unfair on the players at the top of the second half.

[Andrew Stone]

The thing that was not quite right may have been Aaron's own mid-tournament catastrophe. Well, I guess at his level, 4 consecutive draws against players rated 2219-2283 can be described as a catastrophe!

Yeah thanks for that! I was one of the four "catastrophe" oppo's

Most Open's seem to have the same H1vH2 pairings. I'm not sure I've ever played a FIDE rated game against someone within 20 points of me. This is in stark contrast to club chess where you expect an even match. I said to Bob Eames (rated next highest above me) before the London Open that it was impossible for us to play unless we both started off with 5/5. I suppose other unlikely joint leader scenarios would have seen us play.