Wales [Quenched]

[yeti_c]

I think I might finally understand. Thanks yeti!

Clearly this is uncharted territory in terms of the use of the [positions] tag, so it's throwing us all off.

I just made a silly mistake in the previous post is all!!!

But yes it is unchartered territory - as no other map is using similar technology yet.

C.

[MrBenn]

To move away (momentarily) from the discussion about starting positions, I have attempted to work out the probability of starting with shire bonuses from the drop:

My maths might be slightly off, but I've calculated the following:

Code: Select all

Plyrs  Terrs  1 Pair   2 Pairs  3 Pairs  4 Pairs  5 Prs  Total
2/3    10     28.88%   27.40%   16.33%   4.33%    0.20%  77.13%
4       8     35.49%   19.60%   4.78%    0.20%           60.07% 
5       6     31.47%    7.10%   0.27%                    38.85%
6       5     25.16%    2.79%                            27.96%
7/8     4     17.24%    0.62%                            17.86%


[oaktown]

I have a new suggestion for the starts...

1. Pick the four most strategic territories - those that you feel would give the biggest advantage to the player that got them to start - and code one of the two clans in each of them as starting neutrals.
2. The four territories that share a Shire with the above neutrals are regular starts - not in position tags.
3. The remaining 24 territories are split up and as three different start positions.

Why this is better than the earlier suggestions:
• The number of Shires that could be given to a single player are eight in a two player game, zero in a three player game, and twelve in games with 4 or more... this is down from 10, 0, and 12.
• Via the neutrals, you have some control over which Shires will never be held by one player... I would say make them the end Shires, since the combination of instant bonus plus easy to defend/few enemies would be deadly.

You could extend this and pick six or eight Shires to code as half neutral, reducing the advantages further still, but the more neutrals you have the fewer territories everybody starts with... eight might be as far as I'd go, because the 24 remaining terits means only 3 starts in 7/8 player games, 4 starts for 5/6, 6 starts for 4 players, 8 for 3, and 9 for 2.

[e_i_pi]

MrBenn, major apologies for not getting back to you sooner on this one. It's a tricky problem, but it has an easy solution. I've outlined it below in entirety, as this sort of probability is not my forte, and someone with more know-how may find fault that can be rectified. Hopefully the logic is good enough to understand. Someone will either get it and love me, or it will go over everyone's heads

Code: Select all

Problem:

16 pairs of provinces exist
10 red armies exist
10 green armies exist
12 white armies exist

Of all possible distributions of single armies into single provinces, what is the percent chance of:

0 pairs of reds
1 pair of reds
2 pairs of reds
3 pairs of reds
4 pairs of reds
5 pairs of reds

====================================

Let us assume that all 32 armies are dealt out in an undisclosed random fashion, yet we are not given visibility of all 32 provinces.

This is where we require a scientific leap of faith...

Lets say the CC machine only gives us visibility of one province from each pair, in such a way that we can see exactly:

5 pairs with 1 red province and 1 unknown
5 pairs with 1 green province and 1 unknown
6 pairs with 1 white province and 1 unknown

No matter how the armies are distributed, the omniscient CC machine will ALWAYS be able to show us this configuration.  It is simply the nature of the map, and is irrefutable.  If you don't believe me, try it at home.  I'm about 100% sure you can't make a drop that couldn't fit this configuration...

So, our total knowledge of the system is 5 / 5 / 6 visible red / green / white, and 5 / 5 / 6 *invisible* red / green / white.  From this, we can easily and quickly deduce the probability of n pairs of reds where n є {0,1,2,3,4,5}:

P(nR)   = (product sum number of matches x product sum number of non matches) /
              (pool of choices) *
               possible combinations

P(0R)   = (11 x 10 x 9 x 8 x 7)/(16 x 15 x 14 x 13 x 12)
           = 0.10576923076923076923076923076923
           = 10.58%

P(1R)   = (5 x 11 x 10 x 9 x 8)/(16 x 15 x 14 x 13 x 12) x 5 / 1!
           = 0.37774725274725274725274725274725
           = 37.78%

P(2R)   = (5 x 4 x 11 x 10 x 9)/(16 x 15 x 14 x 13 x 12) x (5 x 4 / 2!)
           = 0.37774725274725274725274725274725
           = 37.78%

P(3R)   = (5 x 4 x 3 x 11 x 10)/(16 x 15 x 14 x 13 x 12) x (5 x 4 x 3 / 3!)
           = 0.12591575091575091575091575091575
           = 12.59%

P(4R)   = (5 x 4 x 3 x 2 x 11)/(16 x 15 x 14 x 13 x 12) x (5 x 4 x 3 x 2/ 4!)
           = 0.012591575091575091575091575091575
           = 1.26%

P(5R)   = (5 x 4 x 3 x 2 x 1)/(16 x 15 x 14 x 13 x 12) x (5 x 4 x 3 x 2 x 1/ 5!)
           = 0.00022893772893772893772893772893773
           = 0.02%


This does come to a sum of 100%, and does hold for green as well. The percents will be a tad lower when you start looking at 3 player maps, and significantly lower for 4 player onwards, but I believe 2 player is where the crux of the issue is. If you need further analysis Benn, drop me a PM.

[MrBenn]

It would appear that the maths has turned people off from commenting

My biggest concern is probably that of people getting 3 or more Shire pairs on the drop. I'm not worried about singleton pairs, as I think +1 isn;t a substantial advantage. +2 for two pairs is a little bit more of a concern, but I'm not overly worried about that either...

I'll have a think about creating 3 or 4 starting positions, as these games have the highest probability of multiple pairs...

[yeti_c]

I wouldn't be worried about any games with 4 players - anything 4+ and you have a good drop - you just get beat down by the others...

It's 2/3 player games where the good drop can win you the game before anyone starts that I think you need to worry about.

C.

[MrBenn]

Right... here is my first attempt at starting positions... rather than listing them, I've opted to display them using army placement, which is more visually impactive than an XML extract...

In 2 player games, each player will get a complete set of one colour. 6 territories from the unassigned colour will be shared between the two players, with the remaining territories starting neutral. It is possible for each player to receive a maximum of 3 shire pairs from the drop, but nobody will receive a bonus region from the drop. I think I've caluclated that there is 26.5% of dropping at least one shire pair, 7.5% of getting at least 2, and 3.75% of getting 3 pairs... dramatically lower than the list of probabilities I posted above. (I'm deliberately not posting my working out because I expect I've got it hopefully wrong)

In 3 player games, the armies will be divided up so that each player will get a complete set of one colour. Nobody will start with a shire pair or a region bonus.

4+ player games will be distributed totally randomly, and the starting positions will not be taken into consideration.

Click image to enlarge.

[oaktown]

Looks great Benn. I wouldn't worry too much about somebody scoring a pair in doubles games, because if team 1 gets a pair then his opponent also has one.

I'm trying to wrap my head around where the best places would be for the neutral starts. The South obviously, since it is the smallest region on the board. Where you have the other one makes sense in so far as a neutral in that Shire doesn't set somebody up with a neutral as a defensive blockade, but I also have my eye on Anglesey... with only one bordering Shire, Anglesey/Caernarfon couples to give +2 with only two borders, the only such place on the map.

The region bonuses are blowing my mind - you have to keep considering the Shire bonuses as well... thus Midlands is really a +9 for six territories. I'll stare at that later.

You may want to work on the area between Cardiganshire and Merioneth, as they look like they touch - which I think is not intentional.

[MrBenn]

OK, first thing first... an update

All I've done is make the Cardiganshire / Montgomeryshire /Merioneth border a bit clearer, and moved some neutrals around. (Am I close to a graphics stamp yet?)

I had been contemplating starting one of Anglesey neutral anyway - and it has the added bonus of having an equal distribution of players per region. (Anywhere near a gameplay stamp?)

Click image to enlarge.

Second thing... bonuses...

I haven;t updated the region bonuses since I've added in the mountains... I've run the numbers in to one of the bonus spreadsheets, and come up with the following: (please note the Shire Pairs are taken into consideration in column M)

Click image to enlarge.

I think some of the figures are a bit out, as the region size isn't that much of a factor...
Would it be acceptable to set the region bonus at double the number of Shires? ie North+5; SWest+3; Midlands+3;South+2; English+3? That make North a bit too high.... hmmm....
Any and all thoughts welcome

[oaktown]

All I've done is make the Cardiganshire / Montgomeryshire /Merioneth border a bit clearer, and moved some neutrals around. (Am I close to a graphics stamp yet?)

Gimil usually waits for me to gameplay stamp before he gives his stamp. I'd say it wouldn't hurt to nudge Cardiganshire and Merioneth apart by two more pixels - they're about as close as the top two territories, which do "border" even though they aren't touching. Hmm... might that confuse anyone?

I had been contemplating starting one of Anglesey neutral anyway - and it has the added bonus of having an equal distribution of players per region. (Anywhere near a gameplay stamp?)

I like the current neutral placement. Now, like you said...

Second thing... bonuses...

Right, this is new territory for us since there are so many bonuses in addition to the region bonuses. I think you're on the right track by subtracting the shire bonuses from the calculated bonus. The resulting bonuses look low, but I guess they aren't; the +1 for the big North is really +6. Let's see, this gives you...
North: 10 terits, 4 defenders, +6
Southwest: 6 terits, 4 defenders, +6 (hmm)
Midlands: 6 terits, 6 defenders, +8
South: 4 terits, 4 defenders, +5 (hmm again)
English: 6 terits, 6 defenders, +7.

You may need to think about what factors are most important in this game and re-write the excel spreadsheet accordingly. ??

[gimil]

The 'Cheshire Sir Garelleon' terr name I have a little issue with. 'Garelleon' is a little to worn so some letters are difficult to make out.

[Genghis Khant]

O'r diwedd! Mae rhywun yng nghynllunio map o Gymru. Ansawdd.

...ond, gan ddweud hwnna, dwi'n anghytuno gyda rhai o'r cyfieithu ac mae 'na un neu ddwy camgymeriad sillafu hefyd.

[MrBenn]

O'r diwedd! Mae rhywun yng nghynllunio map o Gymru. Ansawdd.

...ond, gan ddweud hwnna, dwi'n anghytuno gyda rhai o'r cyfieithu ac mae 'na un neu ddwy camgymeriad sillafu hefyd.

"At last! Somebody is making a map of Wales. Quality.
..but I disagree with some of the translations and spelling mistakes."

Is that what you said? Welsh isn't a language I am overly familiar with... I can get as far as Araf, Heddlu, Lifft and Cywm Rhondda...

I'll happily accept any guidance on the translation

[Genghis Khant]

O'r diwedd! Mae rhywun yng nghynllunio map o Gymru. Ansawdd.

...ond, gan ddweud hwnna, dwi'n anghytuno gyda rhai o'r cyfieithu ac mae 'na un neu ddwy camgymeriad sillafu hefyd.

"At last! Somebody is designing a map of Wales. Quality.
..but, having said that, I disagree with some of the translations and there are one or two spelling mistakes as well."

Is that what you said? Welsh isn't a language I am overly familiar with... I can get as far as Araf, Heddlu, codwr and Cwm Rhondda...

I'll happily accept any guidance on the translation

Close, very close. Are you Welsh, or were you educated here?

I'll gladly help you with the translations. What with the two tribes format I reckon you should name the 'continents' after the early medieval kingdoms - Gwynedd, Deheubarth, Powys, Glywysing & Y Gororau (the Marches) - and name the tribes after the cantrefi within them.

In each case north precedes south.

Gwynedd
Anglesey; Môn & Aberffraw
Caernarvonshire; Arfon & Llŷn
Merioneth; Dunoding & Meirionnydd
Denbighshire; Rhôs & Rhufoniog
Flintshire; Tegeingl & Iâl

Deheubarth
Cardiganshire; Ceredigion & Aberteifi
Pembrokeshire; Preseli & Penfro
Carmarthenshire; Dinefwr & Caerfyrddin

Powys
Montgomeryshire; Madog & Gwenwynwyn
Radnorshire; Maelienydd & Elfael
Breconshire; Buellt & Brycheiniog

Glywysing
Glamorgan; Gorfynydd & Penychen
Monmouthshire; Mynwy & Caerwent

Y Gororau
Cheshire; Caerlleon & Maelor
Shropshire; Amwythig & Pengwern
Herefordshire; Henffordd & Ergyng

created by MrBenn = creu gan MrBenn


The poem will take more time, I haven't done any barddoniaeth since I left school.

[Teflon Kris]

Re: Two 'terrets' per territory

Might it be an idea that for territories to attack another territory, both 'terets' should be held?

Or, another idea could be that each territory has a principal town within it which is effectively a seperate territory?

[edbeard]

Or, another idea could be that each territory has a principal town within it which is effectively a seperate territory?

your first idea isn't possible in the current XML so I removed it. This idea could be a good one, though.

Each shire would have a third neutral starting area. This must be held for a shire bonus. It wouldn't be required for a continent bonus. This way, bonuses won't occur on the drop. It adds more strategy too. Do I take over the neutral (2) for a bonus or attack my enemy.

[Genghis Khant]

If you show me how to fit that inside a territory (along with the English translation) the size of the village, I'll put it on

Not my job

"Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch"

The people of the town call it just Llanfair. Check wikipedia.

It's more commonly refered to as "Llanfair P. G.". There must be hundreds of place names starting with "Llanfair" in Wales.

OK, just for fun I'll break it down into it's constituent parts...

Llanfair - church (or parish) of St Mary
pwll - pit (or pool)
gwyn - white
cyll - hazel trees
ger y - by (or near) the
chwyrn - rapid
drobwll - whirlpool
Llandysilio - church (or parish) of St Tysilio
ogof - cave
coch - red

or...

"St Mary's church in the hollow of the white hazels near the rapid whirlpool in the parish of St Tysilio of the red cave"

[yeti_c]

Fantastic - I always wondered what that meant!!

And for more amusement...

http://news.bbc.co.uk/1/hi/wales/7702913.stm

C.

[oaktown]

Or, another idea could be that each territory has a principal town within it which is effectively a seperate territory?

Each shire would have a third neutral starting area. This must be held for a shire bonus. It wouldn't be required for a continent bonus. This way, bonuses won't occur on the drop. It adds more strategy too. Do I take over the neutral (2) for a bonus or attack my enemy.

I think this would make for some really interesting play - probably more so than the current alignment - but explaining it is going to be a bitch. With two territories in each shire, this map already has one feature that will confuse the casual player... the introduction of a third would blow some minds and probably require a graphic showing who can attack what.

Attacks will need to be spelled out: Sub-shire A1 can attack Sub-shire A2, all bordering sub-shires, and Town A (within same shire), but not bordering shire towns. Town A can attack Sub-shire A and Sub-Shire B, but nothing else... correct?

Bonus will also require a complicated explanation on top of this.

[Genghis Khant]

I think two tribes in a territory is perfect. Three would over complicate the gameplay, even if we went with the idea of having a town which could only attack the tribes within its own territory (& vice versa). One of the great things about the two tribes idea is that it's historically accurate. The ancient and medieval Welsh spent more time & effort fighting amongst themselves than repelling the invading Romans, Angles, Normans & English.

As regards the bilingual place names; yes I appreciate that visitors to Wales will see road signs etc in Welsh and English and that people with a basic knowledge of Welsh geography will recognise the names being used, but when it comes to playing the map there are going to be groups of similar looking names in the drop-down menus (Cardiganshire/Carmarthenshire; Sir Gaerfyrddin/Sir Gaernarfon; Meirionnydd/Merioneth) which could be confusing. Besides, it adds a bit more authenticity to be using the names of the old tribal areas (cantrefi) and many of these old names are still in use today (e.g. Llŷn, Arfon, Dinefwr, Ceredigion).

[edbeard]

"complicated" gameplay that leads to a better game

vs

slightly less complicated gameplay that is very limited in the types of games that will work and having the need to structure starting positions which limits the variety of the games themselves

[MrBenn]

Here's an attempt at a version with the altered gameplay, as recently suggested. I really like the way it works, and it does give additional options to make a better game...

I have switched to the Welsh labels as suggested by Genghis Khant, and have used the County towns as the settlements. I've been a bit inconsistent with the use of English/Welsh names for the settlements, so will amend them at some point in the future - but at least they're more-or-less in the right place geographically!

My working image is bigger than the small and large images, so the settlement circles will need to be redrawn at the correct size in a future update. It's worth noting here that the settlements will all start neutral, which will negate the need for complex start-positions.

Click image to enlarge.

[oaktown]

I'd be very careful to get the settlements off of the borders, so it doesn't look as if they can attack out, eg. Cardigan seems to border Preseli/Penfro.

Nice circles.

[yeti_c]

Without the english names - the 2 territories per territory thing doesn't make sense?!

C.

[MrBenn]

Without the english names - the 2 territories per territory thing doesn't make sense?!

C.

Hmmm