Recommend
 
 Thumb up
 Hide
10 Posts

BoardGameGeek» Forums » Board Game Design » Board Game Design

Subject: How to use proportional election method in bg? rss

Your Tags: Add tags
Popular Tags: [View All]
Markku Soikkeli
Finland
Tampere
flag msg tools
mbmbmbmbmb
We just had parliamentary elections in Finland and though the results were depressing as always, I started to think would it be possible to use some kind of d'Hondt-method in a board game. Perhaps it's been already used?

Without any coherent game system in my mind I was thinking about an election game (just saw a preview of Zoocracy!) where the parties (= players) would have to choose what co-op's they will have with other parties to enforce the possibilities of their candidates. But d'Hondt-method takes time to calculate what co-operations may be good for your party. Any suggestions how to simplify the proportional representation in a board game? Or other ways to adapt this method to political games?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Phil Dutré
Belgium
Leuven; Puurs
flag msg tools
designer
Avatar
mbmbmbmbmb
The D'Hondt method (https://en.wikipedia.org/wiki/D%27Hondt_method) indeed is somewhat length to compute a division of seats based on a number of votes (although I remember we had to compute them in school when I was 12 or 13 when elections were covered :-)), but why go through a number of votes first in a game?

Wouldn't it be easier to have votes == seats, and have a straight translation from votes into number of seats?

Another assumption is that the method D'Hondt has the number of parties as a variable. If the number of parties is fixed, you could also use some precomputed table for an approximate distribution of seats - or try to compute the "crossover" numbers at which a seat shifts from one party to the next.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Markku Soikkeli
Finland
Tampere
flag msg tools
mbmbmbmbmb
philip.dutre wrote:

Wouldn't it be easier to have votes == seats, and have a straight translation from votes into number of seats?


Just because of co-ops: part A will have more seats if it has alliance with party B, but party B will have more in alliance with parties C and D. I'm not sure if this kind of co-op option would really mean more suspense... especially with low player count...
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Peter Millen
United Kingdom
Greyabbey
Northern Ireland
flag msg tools
I believe previous games which have tackled PR elections include Dail Eireann and Al Parlamento.

I once made a prototype game based on elections to the Northern Ireland Assembly which uses the Single Transferable Vote (STV) in (at the time) 18 constituencies each electing 6 members.

I used five parties (Sinn Fein - SDLP - Allance - Ulster Unionist - Democratic Unionist) and decided that transferred votes would be split between the parties to the right and left of the transferring party. In addition there was provision for transfers to independent candidates and for 'local pacts' in specified constituencies.

Roughly the main part of the game consisted of players moving round the map and placing potential support (cubes) in constituencies, making nominations and generally boosting standard via debates, manifestoes etc.

A player could request a declaration in a constituency that had not yet elected all six members and which currently had support cubes totalling at least 2 * number of undeclared seats. Various cards could be played to postpone the count, add a last-minute nomination or a late boost in support (more cubes).

The quota for this count was then established as:

Number of support cubes / number of remaining seats : rounded up
eg if there were 8 cubes showing in a constituency with 3 remaining seats the quota is 8/3 = 2.66 = 3.

All the cubes were then thrown into the Tower of Decision and what emerged was the votes counted. Again opportunities to play cards (disqualify votes or recount - back through the tower).

Any party making the quota with a nominee in place got a seat. Then if there were still vacant seats and candidates, surplus votes were redistributed from successful candidates and from the lowest-supported party and checked if anyone now made the quota. Stop the declaration once this procedure fails to produce a returned candidate and return unused cubes to the consituency.

You will see a consituency may have a number of declarations (not a strict simulation) and when all six seats are allocated it is closed off with a 'declared' marker. The game is played to an agreed number of declared constituencies or would not end in a reasonable time.

Alas I only held one playtest to date and would need to put in a lot of work to refine the game but I hope to get back to it oneday.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Jeremy Lennert
United States
California
flag msg tools
designer
Avatar
mbmbmbmbmb
If you're talking about having the players of the game cast actual votes directly, then any sort of proportional representation system is probably inappropriate because the number of voters will be too small.

Also, the properties you want a voting system to have in real life are often different from what you want in a game. Real-life voting systems should be boring, fair, and resistant to manipulation. Voting systems in games should be subtle, tactical, and prone to upset.

Of course, if you're talking about games that tell a story about voting but where the players themselves don't actually vote, then that's completely different.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ian S
United Kingdom
flag msg tools
Avatar
It might be easier to think backwards from the sort of quandaries you want to present i.e. what are the in-game situations you are aiming for, and based on those, how can you model PR to bring out those scenarios?

e.g. if it's simply about getting into power, then the PR element may be better than simply winning the most votes, but also about ensuring that the allying model means you can pal up to get into power, allowing your ally to gain a lesser victory.

So many different ways this could be incorporated. Perhaps Playing Politics would give you some ideas if you could get hold of a copy.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Markku Soikkeli
Finland
Tampere
flag msg tools
mbmbmbmbmb
My first simulation for parliamentary election game (that I was thinking before d'Hondt method) was a design with 5 rows of parties: 3 rows of the big parties (left, center, right), 1 row representing all the old small parties and 1 row representing all the new one-issue-parties (eg the Feminist Party). In every row there would be 4 cards, each card representing 10 seats. So, there's 20 cards for 200 seats of Finnish parliament. The game would be for exactly 3 players, each controlling one of the big parties. The rows are filled from left to right. When a player/party is too arrogant and takes too much risk and fails, the new card goes either to old small party row or to one-issue-party row, depending on mechanics.

If a player gets more than 4 cards to her row she can put her card to the last place in the row in any of other parties - meaning that her own party is so popular it's taking seats from other parties. In addition, this popular party gets a bonus popularity token, and it will be easier to get 'seat cards' in next election.

But how to collect these 'seat cards' and how to regulate the consequences (goverment, ministeries, taxes) between parties with minimal historical accurancy, that's the main problem when trying to choose the suitable mechanics for this game. And that's why I was surprised to see a game design like Zoocracy. The simulation described by mr Millen is also fascinating; I started to think what ideas could I adapt to simulate the Finnish demography.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Kristian Järventaus
Finland
Loviisa
flag msg tools
mbmb
You might be interested in [thread=https://boardgamegeek.com/article/30041984]this recent thread[/thread] about apportioning resources "fairly", which turned to PR pretty quickly.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ludvig Stigsson
Sweden
flag msg tools
mbmbmbmbmb
I think Area Control. The voters give cubes to all partys, then each player (party) use the cubes in votings. Some alliances may form, and they may or may not last the whole game. (If you vote like my party in this voting, I will vote in favor of your party in the next voting). One voting / round, victorypoints goes to the winning side of the vote. Divide vp among the winners each round. Repeat set of rounds. After game end, sum up all colected vp’s to get your partys total. (Then add up every vp’s earned by all players in the game. This will be the new 100%, then do the math to calculate how much % their vp’s make up of that 100%. This will give you the number of next game starting cubes)
Edit: think of Carolus Magnus for inspiration.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Markku Soikkeli
Finland
Tampere
flag msg tools
mbmbmbmbmb
Thanks for new ideas! There's no reason why exactly the d'Hondt method should be used in a bg about party politics. Any simulation of political election might innovate new mechanics (or combine old mechanics) to encourage players to make alliances based on strict rules: backstabbing is ok if you do it like a gentleman.

Instead of d'Hondt method I started to think if it were possible to manipulate the borders of constituencies to get more voters. However, every proportional method have still the same problem with slow calculations. Maybe cube tower ("Tower of Decision"?) is the only easy solution... or to send remainder vote cubes to EU-election (themewise solution).

In a Ludology podcast mr E suggested that some one should do a bg about Venetian elections during the renaissance period. The election system in Venice was extremely complex to prevent bribery and corruption, so it might be a wonderful challenge for players who can see the whole system in wider picture, and manipulate it as they like :-)

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.