Recommend
23 
 Thumb up
 Hide
15 Posts

Commands & Colors: Ancients» Forums » General

Subject: Gallic War Campaign (58-54 B.C) rss

Your Tags: Add tags
Popular Tags: [View All]
Giulio
Italy
Scandiano
RE
flag msg tools
Avatar
mbmbmbmbmb
Dear fellow gamers,
maybe you can find of some interest a new campaign on Gallic War exclusively based on official scenarios. There is a good reason to post this announcement here rather than in the Variants section of this forum. In fact, an essential feature of the campaign is that you are supposed to play the scenarios PRECISELY as they were published. Without any modification. The campaign scoring system will take into account any unbalance in scenarios. Simply, the favored side must earn more banners to win. To balance the campaign, I used the collective wisdom of tenths of players who reported the result of their battles in commandsandcolors.net. This campaign is dedicated to them. I hope you will enjoy it and if you want to know more, read the note that follows.

Designer Notes

Yes I know there are several campaign systems for C&C:A out there and they are very good. But I wanted a different thing. First, I wanted to be able to have a (somehow) competitive confrontation on a series of battles with each player controlling the same side in all battles. Switching sides is a very efficient way of balancing Ancients games but sometimes I like to indulge in the romantic vision of me leading an army, or nation, into a war. Switching sides with my opponent would destroy this vision. Second, I wanted to play the scenarios as they were designed by their creators. Without extra units, extra deployment steps or whatever. The obvious answer for me was to design campaign victory conditions such that the unbalance, possibly presents in every single battle, is composed into a fair campaign. For each scenario, the site commandsandcolors.net reports the fraction of victories of each side from hundredths of individual reports by tenths of players. Thanks to a sophisticated (!) mathematical model I was able to transform these fractions into predictions about the number of medals won by each side. The scoring system of the campaign is based on these predictions.
19 
 Thumb up
1.10
 tip
 Hide
  • [+] Dice rolls
HANJEL T
Spain
flag msg tools
mbmbmbmbmb
Very interesting. I like especially that you represent one faction in the campaign.
Good job.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Mark McG
Australia
Penshurst
NSW
flag msg tools
badge
Avatar
mbmbmbmbmb
why not Gergovia?
https://www.commandsandcolors.net/ancients/maps/77-gallic-wa...

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Giulio
Italy
Scandiano
RE
flag msg tools
Avatar
mbmbmbmbmb
Minedog3 wrote:


Good question Mark. Because it's a scenario which is a bit complicated with several special rules and I wanted the campaign to be accessible also to novice players. If interested, I can include it as optional and add a second scoring table to use when also Gergovia is played.
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Johan Brattstrom
Sweden
flag msg tools
designer
mbmbmbmb
Please do more campaigns like this:). This is awesome!!!!
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Mark McG
Australia
Penshurst
NSW
flag msg tools
badge
Avatar
mbmbmbmbmb
I think this is a great concept, and the fact that it runs over a set of scenarios gives a certain stability to average outcome assumptions. On average the Romans need to win 2 more banners than the Gauls per match for a Minor Roman Victory, and a bit more for a Major Victory.

I'd like to see some extended outcome reports to see how the model works in game play.

Dare I ask for the sophisticated Mathematical model to convert cc.net percentages into banner handicaps? Think it would work with other C&C games (e.g. Napoleonics?)



1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Giulio
Italy
Scandiano
RE
flag msg tools
Avatar
mbmbmbmbmb
If somebody tries the campaign, I'd love to know the scores. I can share with you the results of the final playtest we run a few months ago.

DISCLAIMER: This post is rather technical, which means extremely interesting or amazingly boring, depending on who you are. Consider yourself warned.

These are the expected differences in banners, as predicted by the model. A symbol "-" denotes a fractional difference which is accounted for in the final sum.


% expected
Roman victory banner diff
306 Bibracte 84 +2.9
307 Plain of Alsace 81 +2.4
215 River Sabis 65 +1.2
216 Sotium 52 -
217 Invasion of Britain 81 +2.4
218 River Stour 55 -
219 Foraging party 48 -
Tot: +9.3


While these are the results of our playtest



Giulio Daniele
Romans Barbarians
306 Bibracte 7 4
307 Plain of Alsace 7 1
215 River Sabis 8 4
216 Sotium 6 3
217 Invasion of Britain 7 4
218 River Stour 2 6
219 Foraging party 4 6
Tot: 41 28

Romans banners - Barbarian banners = 13
Expected difference = 9
Final result: +4 excess of banners for the Roman player


It's a Roman minor victory which is, I think, consistent with the essential equivalence in skill between me and Daniele. As you can see, results vary wildly but in the long run, they tend to average out.

The method can for sure be applied to any series of battles of the C&C kind. If I remember well, I calibrated the model the past October. As more battles are reported, the accuracy of the prediction should improve. At some point, I plan to write a short illustrative paper on the methodology. But it's pretty straightforward. For the knowledgeable, suffice to say I used a simple negative binomial model .
4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Matt Crawford
United States
Pacifica
California
flag msg tools
Avatar
mbmbmbmbmb
Fantastic! You did what I have long had it in my mind to do but never could figure out exactly how to do it. Not only the computation, but what scenarios would go well together and make sense to compare. I look forward to reading your write up of your methods.

I’m looking up my solo scores from those battles...

306 Bibracte — Romans 7-5
307 Plain of Alsace — Romans 7-4
215 River Sabin — Romans 8-3
206 Sotium — Romans 8-3
217 Invasion of Britain — Romans 7-4
218 River Stour — Romans 6-5
219 Foraging Party — Romans 6-5

So complete dominance for the Romans in terms of winning individual scenarios. And an overall difference is +20 for the Romans, for a Major Roman Victory. I do remember wondering if the Barbarians would ever win anything.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Giulio
Italy
Scandiano
RE
flag msg tools
Avatar
mbmbmbmbmb
Just to say that a new version with an updated scoring table and an optional, additional, scenario (Gergovia) has been uploaded in the file section.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Marc Gacy
United States
Lafayette
Colorado
flag msg tools
Avatar
mbmbmbmbmb
This is really nice! It is more involved than the quick and dirty method I mentioned in another post: https://boardgamegeek.com/article/29869048#29869048 and therefore more than likely more accurate (you get +9 flags using your analysis and I get +7 for the Gallic War campaign). You have both 306 and 307 having a +2.9 or roughly +3 shift, whereas I top out at +2, which accounts for the difference.

What is your minimum number of scores for you to have confidence in the score for the calculation. I went with 25 for the reasons I stated in the post.

One reason I was initially a little leery of trying to be more accurate is the lack of actual scores. Do you need to assume that a higher win ratio means a greater disparity in scores, or does your formula explicitly account for that with some sort of error calculation?

It is possible (although unlikely) that a scenario with 83% Roman victory has a small point spread (e.g. all the victories are 6-5) and ones with a more reasonable, say 68% Roman Victory, have a really high mean point spread (e.g. all the victories are roughly 6-3). The fact that it is unlikely and the way your example worked out to support the model suggests it's probably not a problem.

Again, great stuff and thanks for doing this!

- Marc
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Stanislav
Belgium
Brussels
flag msg tools
badge
"Glory is fleeting, but obscurity is forever"
Avatar
mbmbmbmbmb
Thanks a lot for this, Giulio. My friend Mikkel and I just played through the campaign. I've posted a campaign AAR here: https://boardgamegeek.com/article/31630033#31630033
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Giulio
Italy
Scandiano
RE
flag msg tools
Avatar
mbmbmbmbmb
Wow! Great battle report! I'm very happy you guys liked the campaign.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Stanislav
Belgium
Brussels
flag msg tools
badge
"Glory is fleeting, but obscurity is forever"
Avatar
mbmbmbmbmb
We thank you for giving us such a great framework.

Do you still have the plans for the short article on the methodology used? I'm digging out my dusty old statistics book and R to see if I can replicate your approach for a Peloponnesian War campaign (and an Alexander the Great one) :-D
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Giulio
Italy
Scandiano
RE
flag msg tools
Avatar
mbmbmbmbmb
Sunnycool wrote:
Do you still have the plans for the short article on the methodology used?


Yes and I think I will illustrate the method proposing three campaigns of 3-5 scenarios each that could be composed in a grand campaign of a dozen scenarios or so. I started working on it already but time is a scarce resource these days.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Stanislav
Belgium
Brussels
flag msg tools
badge
"Glory is fleeting, but obscurity is forever"
Avatar
mbmbmbmbmb
Terrific! It's much appreciated - and much anticipated - but of course there are only so many hours per day, so completely understandable :-)
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls