4NCL Online

Venues, fixtures, teams and related matters.
Mick Norris
Posts: 10310
Joined: Tue Apr 17, 2007 10:12 am
Location: Bolton, Greater Manchester

Re: 4NCL Online

Post by Mick Norris » Thu Aug 13, 2020 10:25 am

Roger Lancaster wrote:
Thu Aug 13, 2020 1:32 am
David Sedgwick wrote:
Wed Aug 12, 2020 6:08 pm

A 17% increase.

My central prediction was a 20% drop. I understand that the organisers thought that it might be greater.

Those of us, including me, who campaigned hard behind the scenes for a change in the ECF and the 4NCl Rules and Regulations can take a quiet satisfaction from the outcome of our efforts.

Those who stormed out in a huff must decide for themselves what if anything they achieved.
Curiously, exactly the opposite outcome to the Junior 4NCL where the organisers seem to have expected an increase but actual entries were well down. I find the contrast puzzling, as I suspect do the organisers. Any theories?
My impression is that there are shed loads of online events for juniors, so that might be a factor, plus the novelty has worn off, and they will be back at school too during season 2 of 4NCL; return of OTB chess might be another

Losing players banned for cheating, plus those dispirited by having played cheats, would be another possible factor
Any postings on here represent my personal views

Roger de Coverly
Posts: 21291
Joined: Tue Apr 15, 2008 2:51 pm

Re: 4NCL Online

Post by Roger de Coverly » Thu Aug 13, 2020 11:03 am

Matthew Turner wrote:
Thu Aug 13, 2020 8:54 am
There is a much more robust appeals procedure for season 2. If a player is flagged by Lichess and doesn't have a high Z score on the Regan test, they will be presumed innocent on appeal.
Presumably the 4NCL still have no say over why lichess might flag an account. If the player is deemed innocent, presumably their old account still stands with its incriminating flag and they would have to set up a new account to continue playing.

John Redmond
Posts: 10
Joined: Sat Sep 01, 2012 6:23 pm

Re: 4NCL Online

Post by John Redmond » Thu Aug 13, 2020 11:29 am

Roger Lancaster wrote:
Thu Aug 13, 2020 1:32 am
David Sedgwick wrote:
Wed Aug 12, 2020 6:08 pm

A 17% increase.

My central prediction was a 20% drop. I understand that the organisers thought that it might be greater.

Those of us, including me, who campaigned hard behind the scenes for a change in the ECF and the 4NCl Rules and Regulations can take a quiet satisfaction from the outcome of our efforts.

Those who stormed out in a huff must decide for themselves what if anything they achieved.
Curiously, exactly the opposite outcome to the Junior 4NCL where the organisers seem to have expected an increase but actual entries were well down. I find the contrast puzzling, as I suspect do the organisers. Any theories?
Some theories:
(a) The 4NCL Online is generally perceived to be a success. The first season was well-organised and enjoyable, albeit there were a few teething problems. To the degree that 'The 4NCL' is a brand, it was enhanced not tarnished by Season 1.
(b) Lichess is an attractive platform. Thanks to recent events more and more players are becoming familiar with it and, generally speaking, it 'suits' the event.
(b) Discussions on the Forum focus so much on cheating they give a distorted picture of the 4NCL Online - they make the event seem endlessly beset and problematic. But this was far from the comfortable and stable round-by-round experience of most players (that I know).
(c) It is (almost) the only show in town. Unless you want to play OTB under 'dystopian conditions' (masks, screens, paranoia) or travel abroad and risk quarantine then there is nothing else.
(d) Some players may have thought that 'normal' OTB chess would resume earlier than now seems likely. Such players may have given the original season a miss calculating they could resume League/4NCL in the Autumn.
(e) Some of the advantages of online chess may have become clearer. As a total experience, The 4NCL Online can't beat the regular 4NCL (I imagine nearly everyone agrees with this) BUT it does have some advantages which will weigh heavily with some: it is is much less of a time-sink and it is much cheaper.
I suppose the real test of the 4NCL Online will be when 'full-blown normality' returns. But we may have to wait a while for that to happen and, for now, this event is on the up.

Caoimhín de Búrca
Posts: 16
Joined: Wed Jun 17, 2020 12:21 am

Re: 4NCL Online

Post by Caoimhín de Búrca » Thu Aug 13, 2020 12:23 pm

Roger Lancaster wrote:
Thu Aug 13, 2020 1:32 am
Curiously, exactly the opposite outcome to the Junior 4NCL where the organisers seem to have expected an increase but actual entries were well down. I find the contrast puzzling, as I suspect do the organisers. Any theories?
For what it's worth, from our point of view as a club that has teams in both, there's a couple of factors I think -

1) The Junior 4NCL (Divs 4/5 for us) was very weak; not much above beginner level on the main.
2) The time control was very short - fine for beginners, but not really conducive to players looking to improve their game
3) We saw a noticeable amount of gamesmanship throughout the Junior 4NCL (offering repeated draws when lost, stalling when lost, distracting chat during games). One team (Woking) even made repeated complaints that people were watching the games online, and their players were asking our players in the chat to tell those watching to go away - ridiculous stuff like that which actually cost us a couple of games. Our players were regularly giving out about it and it took away from their enjoyment of the competition.

So some of our players have jumped to the main tournament as a result. As a general rule, I'm not particularly in favour of junior tournaments and encourage our juniors to play in a main event where possible anyway. So we've lost a junior team but added two senior teams.

Wadih Khoury
Posts: 604
Joined: Sun Jul 12, 2020 8:14 pm

Re: 4NCL Online

Post by Wadih Khoury » Thu Aug 13, 2020 12:40 pm

Usually I recommend juniors to disable the chat when playing games.

Caoimhín de Búrca
Posts: 16
Joined: Wed Jun 17, 2020 12:21 am

Re: 4NCL Online

Post by Caoimhín de Búrca » Thu Aug 13, 2020 1:04 pm

Yep, that was recommended to us alright as a result, though as a lot of the players were new to lichess, we didn't know you could do that at the time.

But yes, would definitely recommend to any junior teams to do that. (Though I can't remember how to do it now! :oops: )

Kevin Thurlow
Posts: 5802
Joined: Wed Apr 30, 2008 12:28 pm

Re: 4NCL Online

Post by Kevin Thurlow » Thu Aug 13, 2020 3:31 pm

"Usually I recommend juniors to disable the chat when playing games."

Yes - probably worth recommending everyone to do that! (Although I can't remember how to do it either)

Roger Lancaster
Posts: 1906
Joined: Tue Mar 17, 2015 2:44 pm

Re: 4NCL Online

Post by Roger Lancaster » Thu Aug 13, 2020 6:28 pm

Caoimhín de Búrca wrote:
Thu Aug 13, 2020 12:23 pm
Roger Lancaster wrote:
Thu Aug 13, 2020 1:32 am
Curiously, exactly the opposite outcome to the Junior 4NCL where the organisers seem to have expected an increase but actual entries were well down. I find the contrast puzzling, as I suspect do the organisers. Any theories?
For what it's worth, from our point of view as a club that has teams in both, there's a couple of factors I think -

1) The Junior 4NCL (Divs 4/5 for us) was very weak; not much above beginner level on the main.
2) The time control was very short - fine for beginners, but not really conducive to players looking to improve their game
3) We saw a noticeable amount of gamesmanship throughout the Junior 4NCL (offering repeated draws when lost, stalling when lost, distracting chat during games). One team (Woking) even made repeated complaints that people were watching the games online, and their players were asking our players in the chat to tell those watching to go away - ridiculous stuff like that which actually cost us a couple of games. Our players were regularly giving out about it and it took away from their enjoyment of the competition.

So some of our players have jumped to the main tournament as a result. As a general rule, I'm not particularly in favour of junior tournaments and encourage our juniors to play in a main event where possible anyway. So we've lost a junior team but added two senior teams.
Coming from a club which had 5 teams in the Online J4NCL [and 3 junior teams in the main event] I'd say you were unlucky in your experience. The top 4 junior divisions were seeded so our teams in those divisions had, almost without exception, a series of fairly even matches. Division 5 was another story altogether. It included our fifth team ["not much above beginner level" would be accurate] whose record was +2 =0 -6. With only one exception, the six losses were total annihilations, all 7-1 or worse. Yet the two wins were both 8-0. In other words, seven of the eight rounds saw complete mis-matches and in some cases any resemblance to chess was entirely fortuitous so I'm at a bit of a loss to explain how those juniors enjoyed it. But they're back so evidently they did. I suppose it just goes to show that chess can be enjoyed at many levels.

Nick Grey
Posts: 1838
Joined: Thu Dec 01, 2011 12:16 am

Re: 4NCL Online

Post by Nick Grey » Thu Aug 13, 2020 8:01 pm

with 1,000m committed to catch up education funding Online J4NCL ought to be down.
Overall numbers good. Well done.

Caoimhín de Búrca
Posts: 16
Joined: Wed Jun 17, 2020 12:21 am

Re: 4NCL Online

Post by Caoimhín de Búrca » Thu Aug 13, 2020 10:31 pm

Roger Lancaster wrote:
Thu Aug 13, 2020 6:28 pm
Caoimhín de Búrca wrote:
Thu Aug 13, 2020 12:23 pm
Roger Lancaster wrote:
Thu Aug 13, 2020 1:32 am
Curiously, exactly the opposite outcome to the Junior 4NCL where the organisers seem to have expected an increase but actual entries were well down. I find the contrast puzzling, as I suspect do the organisers. Any theories?
For what it's worth, from our point of view as a club that has teams in both, there's a couple of factors I think -

1) The Junior 4NCL (Divs 4/5 for us) was very weak; not much above beginner level on the main.
2) The time control was very short - fine for beginners, but not really conducive to players looking to improve their game
3) We saw a noticeable amount of gamesmanship throughout the Junior 4NCL (offering repeated draws when lost, stalling when lost, distracting chat during games). One team (Woking) even made repeated complaints that people were watching the games online, and their players were asking our players in the chat to tell those watching to go away - ridiculous stuff like that which actually cost us a couple of games. Our players were regularly giving out about it and it took away from their enjoyment of the competition.

So some of our players have jumped to the main tournament as a result. As a general rule, I'm not particularly in favour of junior tournaments and encourage our juniors to play in a main event where possible anyway. So we've lost a junior team but added two senior teams.
Coming from a club which had 5 teams in the Online J4NCL [and 3 junior teams in the main event] I'd say you were unlucky in your experience. The top 4 junior divisions were seeded so our teams in those divisions had, almost without exception, a series of fairly even matches. Division 5 was another story altogether. It included our fifth team ["not much above beginner level" would be accurate] whose record was +2 =0 -6. With only one exception, the six losses were total annihilations, all 7-1 or worse. Yet the two wins were both 8-0. In other words, seven of the eight rounds saw complete mis-matches and in some cases any resemblance to chess was entirely fortuitous so I'm at a bit of a loss to explain how those juniors enjoyed it. But they're back so evidently they did. I suppose it just goes to show that chess can be enjoyed at many levels.
It's possible, but I'm not sure. Apart from the divisional final (which we lost on tie-break to a decent side), our Div 4 team won all 9 of our matches despite being bottom seed. We scored 62/72 in terms of game points - an average of 7-1 - even with a board 4 who'd never played a rated game before. I was watching a fair few online and thought the standard was objectively low; a lot of basic errors and easy wins. Some of the points we dropped were pure carelessness as the games just weren't challenging (plus one or two of our players getting put off by the gamesmanship mentioned - which they should have ignored of course, but they're juniors) So it spanned the whole of Div 4.

Maybe we were an exception. Irish ratings are stronger than the same FIDE/ECU rating - 1600 Irish is about 1700 FIDE for example - and that would have us seeded lower than we maybe deserved. You can see the Irish sides were overrepresented in the knockouts in general and I'd say it's because we were all seeded too low. The organisers can't really go taking every quirk like that into account of course.

That's just our experience in terms of why we've dropped a junior team but added a senior team. I don't mean it to be an absolute fact of course; other clubs will have their own reasons and the overall picture is a tapestry of the lot.

User avatar
Adam Raoof
Posts: 2720
Joined: Sat Oct 04, 2008 4:16 pm
Location: NW4 4UY
Contact:

Re: 4NCL Online

Post by Adam Raoof » Thu Aug 13, 2020 10:42 pm

Not quite the only show in town!

Go to www.chessengland.com

You can find all my tournaments on Tornelo here - https://tornelo.com/chess/orgs/chess-england

Adam
John Redmond wrote:
Thu Aug 13, 2020 11:29 am
Roger Lancaster wrote:
Thu Aug 13, 2020 1:32 am
David Sedgwick wrote:
Wed Aug 12, 2020 6:08 pm

A 17% increase.

My central prediction was a 20% drop. I understand that the organisers thought that it might be greater.

Those of us, including me, who campaigned hard behind the scenes for a change in the ECF and the 4NCl Rules and Regulations can take a quiet satisfaction from the outcome of our efforts.

Those who stormed out in a huff must decide for themselves what if anything they achieved.
Curiously, exactly the opposite outcome to the Junior 4NCL where the organisers seem to have expected an increase but actual entries were well down. I find the contrast puzzling, as I suspect do the organisers. Any theories?
Some theories:
(a) The 4NCL Online is generally perceived to be a success. The first season was well-organised and enjoyable, albeit there were a few teething problems. To the degree that 'The 4NCL' is a brand, it was enhanced not tarnished by Season 1.
(b) Lichess is an attractive platform. Thanks to recent events more and more players are becoming familiar with it and, generally speaking, it 'suits' the event.
(b) Discussions on the Forum focus so much on cheating they give a distorted picture of the 4NCL Online - they make the event seem endlessly beset and problematic. But this was far from the comfortable and stable round-by-round experience of most players (that I know).
(c) It is (almost) the only show in town. Unless you want to play OTB under 'dystopian conditions' (masks, screens, paranoia) or travel abroad and risk quarantine then there is nothing else.
(d) Some players may have thought that 'normal' OTB chess would resume earlier than now seems likely. Such players may have given the original season a miss calculating they could resume League/4NCL in the Autumn.
(e) Some of the advantages of online chess may have become clearer. As a total experience, The 4NCL Online can't beat the regular 4NCL (I imagine nearly everyone agrees with this) BUT it does have some advantages which will weigh heavily with some: it is is much less of a time-sink and it is much cheaper.
I suppose the real test of the 4NCL Online will be when 'full-blown normality' returns. But we may have to wait a while for that to happen and, for now, this event is on the up.
Adam Raoof IA, IO
Chess England Events - https://chessengland.com/
The Chess Circuit - https://chesscircuit.substack.com/
Don’t stop playing chess!

Roger Lancaster
Posts: 1906
Joined: Tue Mar 17, 2015 2:44 pm

Re: 4NCL Online

Post by Roger Lancaster » Fri Aug 14, 2020 9:00 am

Incidentally, I rather liked this

"4NCL Online Format Regulations

"4. Organisation of the Divisions 1. Divisions 1, 2, 3 and 4 will each compromise four groups of eight teams …"

I take it that further compromising arose over the proof-reading budget rather than this being intended as an omen!

Roger Lancaster
Posts: 1906
Joined: Tue Mar 17, 2015 2:44 pm

Re: 4NCL Online

Post by Roger Lancaster » Fri Aug 14, 2020 7:54 pm

Alex Holowczak's recent article, published in CHESS and elsewhere, was well written and has rightly attracted near-universal praise on this forum so I hope no-one, least of all Alex, will mind if I point out something of a small problem.

Alex uses a graph which overlays performances in over-the-board 4NCL and Online 4NCL, assuming that both conform to a Normal distribution or something closely approximating to it and that the two sets of data have identical means and identical standard distributions. There's a limit to how precise I can be without access to the raw data which Alex has grouped, entirely reasonably, into histograms for presentational purposes - the simplest assumption for analysis purposes is that data is evenly distributed within each histogram but this is an approximation only. There are some other statistical caveats which I'm happy to discuss privately but which likely won't be of much interest to those none too fond of statistics.

I won't dwell on the respective means because my calculation is that the online mean is somewhere between -0.1 and -0.2 which actually reinforces, rather than detracts from, Alex's argument - although only slightly. My first issue is with the standard deviations. For the over-the-board graph, 68% of the data falls in the 1>z>-1 range which is exactly what one might expect from a Normal distribution. However, in the case of the online graph, only 58% [plus or minus a small figure, perhaps 0.5%, to allow for inexact data as referred to earlier] falls within that 1>z>-1 range.

That means that, if the online graph indeed follows a Normal distribution as Alex seems to suggest, it's standard distribution is not z. So let's call it s. Reference to statistical tables reveals that, in the case of Normal distributions, 57.6% of the data falls in the 0.8>s>-0.8 range. That gives us the equation z= {0.8 x s} or, if you prefer, s = {1.25 x z}. That would mean that, where for example the graph purports to show an online player's performance at z=4, the correct figure should be s=3.2. In itself, that doesn't nullify Alex's contention but it does appreciably weaken it.

My second issue is with the assumption that the online data can safely be regarded as constituting a Normal distribution. If no-one cheated, I'd have no problem with this. But Alex is clear that cheating occurs [and I disagree only as regards the frequency of cheating] in which case the assumption becomes highly questionable.

At one extreme, if the weakest 5% were to cheat, this would result in the truncation - from what would otherwise have been a Normal distribution - of all data occurring below -1.7 times the standard deviation and its relocation to a point above +3 times the standard deviation. What's left plainly wouldn't a Normal distribution. In principle, that applies [but in different ways] no matter who cheats. There are ways of attempting to adjust for this but they beyond the scope of a short post - and very probably also beyond my capabilities, not having studied the subject for 50 years, as a statistician. What this second issue questions is whether a Normal distribution is likely to give a reliable result when it is most unlikely that the data actually conforms to a Normal distribution.

John McKenna

Re: 4NCL Online

Post by John McKenna » Fri Aug 14, 2020 9:23 pm

"... whether a Normal distribution is likely to give a reliable result when it is most unlikely that the data actually conforms to a Normal distribution."

I believe whether or not a data set conforms to a Normal distribution can be tested for.

Roger Lancaster
Posts: 1906
Joined: Tue Mar 17, 2015 2:44 pm

Re: 4NCL Online

Post by Roger Lancaster » Sat Aug 15, 2020 8:10 am

John McKenna wrote:
Fri Aug 14, 2020 9:23 pm
"... whether a Normal distribution is likely to give a reliable result when it is most unlikely that the data actually conforms to a Normal distribution."

I believe whether or not a data set conforms to a Normal distribution can be tested for.

The extract reproduced by John was intended as a maxim of general application. In the particular case of the online data, it's part of my case that it patently doesn't conform.

Post Reply