4NCL North 2015-16

[David Pardoe]

Mike Truran is currently consulting captains over a two-pool model for Div 3N this season, given the adventitious number of teams. Spirit of Atticus have agreed, of course. I hope there's general assent. The model isn't perfect: an APA is better; a Swiss, worse. But the two-pool model does make for lively competition. Let's do it

Who is this `interesting ` for....David ?
Certainly not the second tier group who might feel they are `second class rejects`, doomed to play the `also rans` and rag end teams...
They might rightly feel condemned to a Div 4 status, left to play boring meaningless matches against the other `rejects`..
How very depressing....could sink the league if we`re not careful..

Meanwhile, those in the top pool will correctly feel elated and fired up to play the other top dogs...

At 11 rounds, we are almost playing an APA if we go for the Swiss, and all teams can look forward to some challenging and interesting matches.. surely that's what its all about.. The top teams have 11 rounds to show there metal....
The Swiss in this situation gives the league a `cup atmosphere` ...and I`m really not up for this business of knowing everything months in advance.
The stooges will then do there homework, prep up on the opposition and..hey...how boring.

Lets have some real cup clashes, where surprises can happen...yes, one or two teams might get battered occasionally ...but otherwise, much more interesting I think.

[Roger de Coverly]

Certainly not the second tier group who might feel they are `second class rejects`, doomed to play the `also rans` and rag end teams...

I rather thought the idea was to have two pools of approximately equal nominal strength. That at least is a means of preventing first team v second team clashes by the simple device of putting the teams in opposite pools. After round 7, the top 4 of each group merge as do the bottom 4. So it's only if a team scores badly in the first 7 rounds, that it becomes an "also ran".

By the very nature of the construction of the groups, there are going to be clashes between teams looking to be promoted and rag end teams in the first seven rounds.

[MartinCarpenter]

I'm just very happy there's no triangles! Swiss or pools should work out pretty similar in the end.

It won't prevent 1st vs 2nd team clashes - with how the squads you'd expect MM2 at least to have a very good chance of ending in the promotion pool. Maybe DCA B. Some argument for putting MM1/2 and DCA A/B in the same pools actually.

There really isn't much of a rag end to the division either. A handful of weak teams who'll play each other regardless, but only 4 or so actively so I'd guess.

[Mick Norris]

I'm just very happy there's no triangles! Swiss or pools should work out pretty similar in the end.

It won't prevent 1st vs 2nd team clashes - with how the squads you'd expect MM2 at least to have a very good chance of ending in the promotion pool. Maybe DCA B. Some argument for putting MM1/2 and DCA A/B in the same pools actually.

There really isn't much of a rag end to the division either. A handful of weak teams who'll play each other regardless, but only 4 or so actively so I'd guess.

I'd agree with all of that, including whether to put those teams in the same pools, and have them play each other in round 1? If not, they might end up playing each other in round 8 anyway

Personally, a mix of team strengths to play against sounds good, and then the last 4 rounds involves our 3rd team doing our best in the "trying not to finish last" pool

[David Pardoe]

Certainly not the second tier group who might feel they are `second class rejects`, doomed to play the `also rans` and rag end teams...

I rather thought the idea was to have two pools of approximately equal nominal strength. That at least is a means of preventing first team v second team clashes by the simple device of putting the teams in opposite pools. After round 7, the top 4 of each group merge as do the bottom 4. So it's only if a team scores badly in the first 7 rounds, that it becomes an "also ran".

By the very nature of the construction of the groups, there are going to be clashes between teams looking to be promoted and rag end teams in the first seven rounds.[/quote/]
********************************************************************************************************************************************************


Thanks for your comments Roger...
I still think, with 16 teams and 11 rounds, there`s no need to mess around creating pools...a Swiss would do the job and create some interesting pairings throughout the duration.... and no dogs breakfast pools, which I think could be a turn-off...
Yes, for Div 1 super league, I can see a certain logic in splitting things up... and the form book might be rather more precise in the lofty towers of 2450 - 2700 chess...
Part of the fun of this league is turning up on Sunday morning to see who your opponents will be.. team and players..

[David Robertson]

Some argument for putting MM1/2 and DCA A/B in the same pools actually

Yes, I arrived at this point last night, thinking how the Pools might shape up. First though: what principles guide the allocation to Pools A & B? Criteria exist, after a fashion, for Div 1 & 2. But there's no precedent here. Best we think it through between us now because, if we leave it to the Arbiters, you can be sure they'll botch it. So...

principles

1. equalise Pool A & B
2. minimise matches within squads
3. prefer regional diversity

Only the third needs explanation: that is, where a choice of either Pool exists, place a team in the Pool with fewer co-locators

method of allocation

Pair:
B x3 with MM x3
C x2 with SoA x2
J with HC

and, for the moment, pending news on squads
A with BB
MR with 3Cs

allocate to Pool

Pool A: MM1; MM2; B3; C2; SoA3; J; BB; MR

Pool B: B1; B2; MM3; C3; SoA2, HC; A; 3Cs

From these Pools, you can 'predict' Pools C & D from what we know to date, as a means of cross-checking the balance between initial Pools. My guess/intuition is that the balance looks about right; and the Pools will involve a right old scrap, as in Div 2

[Mick Norris]

principles

1. equalise Pool A & B
2. minimise matches within squads
3. prefer regional diversity

Only the third needs explanation: that is, where a choice of either Pool exists, place a team in the Pool with fewer co-locators

Good spot, David, the regional bit would be good, gives more of a chance for players to play new opponents (which is why it is good to see Alba, Manx Ravens and Broadland join)

[Bob Clark]

Best we think it through between us now because, if we leave it to the Arbiters, you can be sure they'll botch it

In my experience the 4NCL arbiters are the best in the country, I have every confidence that they will split the pools fairly.

More importantly the first criteria for splitting the pools is where they finished in the previous season.
The top eight from last season were including Broadland is as follows:
1) Cheddleton 2
2) Broadland Bitterns
3) MM 2
4) SOA B
5) Bradford A
6) Holmes Chapel
7) MM 1
8 ) Bradford B

These teams should be allocated to the pools first. The remaining four teams from last season in order:
Jorvik, MM3, Bradford C and Cheddleton 3 can then be allocated.

The four new teams: SOA C, Alba, 3C's 2 and Manx can then complete the groups.

This follows the principle established a couple of years ago when the early pairings were based on the finishing position in the previous season.

[David Pardoe]

All this guesswork to try to predict team strength, etc...before we have any idea who might be playing...?
Mystic Meg required...

Note that MM1 finished 7th and MM2 finished 3rd...last season...comparitavely?
And looking at last years results, did NE2 (a now extinct animal, sadly), not pull a shock win against potential table toppers Bradford A..... in the first rounds....
All this forecasting and `fixing` is a nonsense...

In practice, taking account of which players might actually be available on any given weekend, etc....not to mention other gremlins which might occur...this guesswork is not helpful?
Yes, by all means, lets have those chats down the pubs and eating houses on the Saturday night.... but leave it there and look forward to our Arbitors pairings on the Sunday morning....that`s my take.

As I`ve said, the margins between the main body of teams (ie, teams ranked 4 to 12), is probably fairly small....

Lets stick with the Swiss....at least that takes some account of actual performance and results...with some conventional `arranged` pairing strategies in early rounds..
Incidentally, there are no duffer teams in the 4NCL North.... just winners and runners-up!!

[David Robertson]

In my experience the 4NCL arbiters are the best in the country

Which is the problem. For those of us with broader experience, the evidence does not lend confidence

they will split the pools fairly

They will - as they see it. No one claims Arbiters are corrupt; merely that they make decisions too often at variance with the playing community's reasonable wishes

More importantly the first criteria for splitting the pools is where they finished in the previous season...This follows the principle established a couple of years ago when the early pairings were based on the finishing position in the previous season

Why "more importantly"? In a Pool system, the first criterion should be to avoid strength imbalances between Pools. The mechanistic, 'finishing position', formula is a regular cause for disquiet in both Div 1 & 2. It's tolerated because players and captains are broadly familiar with each other; and no other criteria are sought.

But this is not the case, as far as comments in here suggest, for Div 3N. Some are proposing minimal contact between teams in the same squad; others, that regional diversity should be pursued in an otherwise pretty closed playing community. Hence, for example, 3Cs should be pooled separately from MM; and presumably, by the same token, Jorvik separately from Bradford. If one uses historic data mechanistically, these additional, and quite sensible wishes get trampled over. Instead, I'm offering a player-led consensus-building model for Pool allocations.

[MartinCarpenter]

Seems a worthy goal

With Jorvik its actually mostly been DCA B that's worth dodging - they've always had a few York players and even been majority York based at times.

Jorvik - DCA A doesn't actually feature many overly repeated clashes. No real evening league overlap, several players on both teams are mostly outside the Yorkshire league and those that are wouldn't expect to meet due to being on quite different boards. Ditto for C I guess.

[Mick Norris]

The top eight from last season were including Broadland is as follows:
1) Cheddleton 2
2) Broadland Bitterns
3) MM 2
4) SOA B
5) Bradford A
6) Holmes Chapel
7) MM 1
8 ) Bradford B

These teams should be allocated to the pools first. The remaining four teams from last season in order:
Jorvik, MM3, Bradford C and Cheddleton 3 can then be allocated.

The four new teams: SOA C, Alba, 3C's 2 and Manx can then complete the groups.

Presumably 1,4,5 & 8 and 2,3,6&7?

Pool A - Ch2, SoA B, Br A&B
Pool B - BB, MM2, HC, MM1

Not much different then - but if you then mechanically add 9 (Jorvik) & 12 (Ch 3) to A you get both Cheddleton teams in A, but if you add them to B, you get 3 Bradford teams in A

Surely better to expect the 4 weakest teams to be SoA C, MM3, Br C & Ch 3? And "group" Jorvik with 3Cs 2, Manx & Alba?

Alternatively, you could easily follow precedent and treat BB as new to Div 3 N, place them alongside 3Cs 2 (who do have a track record in Div 3 S, albeit not last season), Manx & Alba, bumping Jorvik up the list into the top 8, giving you:

Pool A - Ch2, Br A, HC, J
Pool B - MM2, SoA B, MM1, Br B

I prefer David's approach to be honest

This follows the principle established a couple of years ago when the early pairings were based on the finishing position in the previous season.

Which I thought worked badly when new teams were introduced to the league

[MartinCarpenter]

Any seeding system which put Jorvik above Broadlands would be very strange I don't honestly think trying to carefully balance strength in the pools would be so crucial as the teams are so well matched anyway, and some are quite volatile in terms of strength/weekend.

[Bob Clark]

The top eight from last season were including Broadland is as follows:
1) Cheddleton 2
2) Broadland Bitterns
3) MM 2
4) SOA B
5) Bradford A
6) Holmes Chapel
7) MM 1
8 ) Bradford B

These teams should be allocated to the pools first. The remaining four teams from last season in order:
Jorvik, MM3, Bradford C and Cheddleton 3 can then be allocated.

The four new teams: SOA C, Alba, 3C's 2 and Manx can then complete the groups.

Presumably 1,4,5 & 8 and 2,3,6&7?

Pool A - Ch2, SoA B, Br A&B
Pool B - BB, MM2, HC, MM1

Not necessarily and why I didn't attempt to set up the groups merely the principle.

You could for example place positions 1,3,6 and 8 in one group and 2,4,5 and 7 in another. Which would be fair.
You could place 1,2,7,8 in one group and 3,4,5,6 in the other

The principle remains though that teams should be pooled based on their previous seasons results.

Note last season Guildford 1 and Cheddleton 1 (the best two teams) ended up in the same division 1 pool because previous years results were used.

[Mick Norris]

The very strongest teams aren't the issue, as they will all have to play each other anyway, however nice a round 11 match up between the best 2 teams would be

The issue is to try and make sure that the "best" 8 teams go into the Promotion pool, so the placing of teams 5-12, say, is probably what matters - you could go with 1,2,7,8,9,10 and 3,4,5,6,11,12 say

Presumably, someone can work out from Div 1 2014-15, and from the seedings from the final places in 2013-14, what was actually applied in practice

However, back to the fundamental point, I think maybe half the teams are playing for fun not for promotion, so David's criteria might improve the experience

As Martin says, not much between lots of the teams in practice