An improved Preferences method

This is the home for suggestions for site improvements, changes to house rules, and new variants.
Forum rules
It's okay to suggest new rules variants in this forum, but proposing new *maps* should be done in the linked "New Map Variant Proposals & Voting" subforum.

An improved Preferences method

Postby Mr.E » 06 Apr 2019, 14:20

I've made this suggestion before but it is hidden away in a bigger discussion, so I'm making it again as an alternative to the FCFS proposal made by NoPunIn10Did: The Malta Method. This isn't because the Malta Method is a bad suggestion, but because FCFS is a poor idea in itself.

Any method based on when a player joins a game is poor. It means that the game's creator has ultimate choice, and that anyone who joins after her is restricted in their choices. Why should this be? Should it be important that when I join a game should restrict my options over which power I play?

My argument is no based on the fact that this is, at best, naturally unbalanced and, at worst, it encourages people to create new games because they can't play the power they want in existing games, and this means a lot of games sat around, not starting, because nobody wants to play the powers that are left.

The Preferences method achieves exactly the same as FCFS... or it could do... without the bias based on when players join the game. FCFS can be made redundant.

So here it is:
    > When players CONFIRM they are playing the game, they are presented with a list of all the powers (for Versailles, all principal powers). They get to order the powers by preference. They can choose to order all powers rather than just selecting three.
    > Players can make fewer choices than the full list if they want, although this has implications (see below). If it is that a player wants to not play a power, though, then placing that power in last place in the selection is important, so if this is what you want, you will need to select all powers in order.
    > A power cannot be selected more than once in the preferences list but changes can be made by selecting alternative powers earlier in the list.
    > When your selections are ready, players CONFIRM.
    > An option would be available for a player to create an empty preferences list by selecting CONFIRM but a warning should pop up if not all positions on the list have been filled, requiring a second confirmation.
At this point, nobody is penalised by when they joined the game, and everyone has as good a chance of playing the power they want as everyone else... potentially. The next part of the process is automatic, requiring no player interaction.
    Pref 1 (a): Any power that has been selected as a first preference which has only been selected as a first preference by one player is allocated to that player. That power is removed from further consideration and that player's further preferences are ignored.
    Pref 1 (b): Any power that has been selected as a first preference which has been selected as a first preference by multiple players is allocated to the player who is the lowest rated. That power is removed from further consideration and that player's further preferences are ignored.
    Pref 2 (a): Any players who haven't yet been allocated powers will have their lists re-prioritised so that allocated powers are removed from their list. In practice, if they didn't get their first choice, and their second choice has been allocated to another player, their third choice tops their list.
    Pref 2 (b): Follow the process for Pref 1 (a) above.
    Pref 2 (c): Follow the process for Pref 1 (b) above.
This process is repeated until all powers have been allocated. However, some players may not have filled all places on their allocation lists - so what happens to them?
    > If a player exhausts her preferences list and remains without a power allocated to her, then she receives the remaining power.
    > If multiple players exhaust their preferences list as above then remaining powers are allocated randomly.

In this process, where two or more players select the same power in the same part of the process (as in Pref 1 (b)) powers are allocated to the lowesr rated player of the group. This isn't absolutely necessary - it could be the highest rated player that gets the preference. This means that the lowest rated player that expresses a preference will always be allocated her choice.

This system could be used for tournaments as well, with each player being told to select just one power and this would be the power he was allocated.

This would take away the inpatient complaints of players who draw the same power consecutively under the Random system (well, almost certainly) and give players the chance to play different powers. It isn't based on a FCFS system, divorcing completely when a player joins the game from the allocation system.

By placing this selection process in the confirmation stage, it means it could replace the confirmation stage. You'd be confirming by making a preferences selection (or confirming that you have completed your list).

Here's a few examples. In each example, ties are broken by the lowest rated player given priority, and player order (1, 2, 3 etc) lists players by rating.
Code: Select all
PREFERENCES
PLAYER    PREF 1    PREF 2    PREF 3    PREF 4    PREF 5    PREF 6    PREF 7
Player 1  Turkey    France    Russia    Germany   Austria   Italy     England
Player 2  Turkey    Germany   Austria   -         -         -         -
Player 3  Russia    Italy     Turkey    France    England   Germany   Austria
Player 4  Italy     Turkey    England   Russia    -         -         -
Player 5  Italy     Turkey    England   Austria   Germany   Russia    France
Player 6  Germany   France    Italy     Austria   Turkey    England   Russia
Player 7  France    -         -         -         -         -         -

RESULTS   ROUND 1   ROUND 2   COMMENT
Player 1  -         Austria   In R1, priority given to P2 for Turkey; in R2, France, Russia, Germany allocated in R1, so Austria                                      becomes next choice.
Player 2  Turkey
Player 3  Russia   
Player 4  -         England   In R1, priority give to P5 for Italy; in R2, Turkey allocated in R1 so England becomes next choice.
Player 5  Italy
Player 6  Germany   
Player 7  France


Code: Select all
PREFERENCES
PLAYER    PREF 1    PREF 2    PREF 3    PREF 4    PREF 5    PREF 6    PREF 7
Player 1  England   Italy     -         -         -         -         -
Player 2  France    -         -         -         -         -         -
Player 3  Italy     England   France    Austria   Germany   Russia    -
Player 4  France    England   -         -         -         -         -
Player 5  Austria   Russia    Turkey    Germany   -         -         -
Player 6  Russia    Italy     -         -         -         -         -
Player 7  France    Italy     -         -         -         -         -

RESULTS   ROUND 1   ROUND 2   COMMENTS
Player 1  England
Player 2  -         Turkey    In R1, priority given to P7 for France; player had exhausted preference list, so randomly allocated from                                Turkey/Germany.
Player 3  Italy
Player 4  -         Germany   In R1, priority given to P7 for France; in R2, England had been allocated in R1 so player had exhausted                                preference list, so randomly allocated from Turkey/Germany.
Player 5  Austria
Player 6  Russia
Player 7  France


Code: Select all
PREFERENCES
PLAYER    PREF 1    PREF 2    PREF 3    PREF 4    PREF 5    PREF 6    PREF 7
Player 1  Turkey    Austria   France    Italy     Russia    -         -
Player 2  Germany   England   Russia    -         -         -         -
Player 3  Austria   Russia    England   -         -         -         -
Player 4  Italy     England   -         -         -         -         -
Player 5  No preferences stated
Player 6  Austria   Turkey    England   -         -         -         -
Player 7  France    Turkey    Italy     Germany   England

RESULTS   ROUND 1   ROUND 2   ROUND 3   COMMENTS
Player 1  Turkey
Player 2  Germany
Player 3  -         Russia              In R1, priority given to P6 for Austria.
Player 4  Italy
Player 5  -         -         England   Having left preferences blank, England was allocated as the only power remaining after R2.
Player 6  Austria
Player 7  France


Code: Select all
PREFERENCES
PLAYER    PREF 1    PREF 2    PREF 3    PREF 4    PREF 5    PREF 6    PREF 7
Player 1  France    -         -         -         -         -         -
Player 2  France    Italy     Germany   Russia    Austria   -         -
Player 3  France    Russia    Germany   Turkey    -         -         -
Player 4  No preferences stated
Player 5  France    England   Russia    Germany   Italy     Austria   Turkey
Player 6  France    England   -         -         -         -         -
Player 7  France    Italy     Germany   Russia    Austria

RESULTS   ROUND 1   ROUND 2   ROUND 3   ROUND 4
Player 1  -         -         -         Turkey
Player 2  -         Italy
Player 3  -         Russia
Player 4  -         -         -         Austria
Player 5  -         -         Germany
Player 6  -         England
Player 7  France

Comments
In this final game all players except for the undecided P4, chose France as their first preference. Given that the lowers rated player breaks tie, P7 was allocated France and everyone else moved on to their second choice.
In Round 2, P2 & P3 got their choice as the only players who had Italy and Russia as the top of their list; P6 was allocated England, prioritised over P5 as the lowest-rated of the two.
In Round 3, with France, England & Russia already allocated, Germany was P5's next choice. As P1 had exhausted his preference list in R1, and P4 had no preferences entered, Germany was allocated to P5.
In Round 4 that left P1 & P4 without an allocation, and Turkey & Austria to be allocated. Neither player had any preferences to refer to, so the powers were allocated randomly.
Respect neither opinions nor beliefs; only respect the person and the right to express them.
Play by the rules but be ferocious.
User avatar
Mr.E
Premium Member
 
Posts: 160
Joined: 20 Feb 2017, 09:27
Location: Yorkshire
Class: Diplomat
Standard rating: 974
All-game rating: 974
Timezone: GMT

Re: An improved Preferences method

Postby NoPunIn10Did » 08 Apr 2019, 18:15

Have you read this thread that Dipsy posted describing how the existing Preferences system works?
https://www.playdiplomacy.com/forum/viewtopic.php?f=564&t=59548

I think the only fundamental difference between what you're suggesting and how Preferences already works today is the number of countries that a player lists.

Listing seven different dropdown boxes (rather than the current three) could get cumbersome interface-wise. I'd also be curious how often somebody really makes a distinction between their 4th, 5th, 6th, and 7th choices, especially considering just how many people will list the same country in multiple boxes (though doing so has no effect on their chance to be assigned that country).

In my time as a GM, it's been my experience that only a few players really care past the top 2-3 picks.
NoPunIn10Did
Moderator (Forums only)

Variant GM, Designer & Collaborator
User avatar
NoPunIn10Did
Premium Member
 
Posts: 2031
Joined: 17 Aug 2011, 00:17
Location: North Carolina
Class: Ambassador
Standard rating: 1000
All-game rating: 1515
Timezone: GMT

Re: An improved Preferences method

Postby super_dipsy » 08 Apr 2019, 21:58

We are not having preferences with 7 country drop downs. That is just plain silly. There have been suggestions about possibly changing preferences, but they are about execution of the preferences resolution and nothing to do with having more than 3 choices.

i think what people have to remember is that perfection comes second to usability. The key is to balance the function with ease of understanding and usability.
User avatar
super_dipsy
Premium Member
 
Posts: 11847
Joined: 04 Nov 2009, 17:43
Class: Ambassador
Standard rating: 1000
All-game rating: 941
Timezone: GMT

Re: An improved Preferences method

Postby Mr.E » 09 Apr 2019, 09:26

Yes I read what Dipsy said about how Prefs worked. What I suggested was to minimise the random choice part of it. Frankly, if Prefs reverts to random so quickly, it isn't really preferences.

I understand what Dipsy's saying about usability. If what I've suggested isn't usable, fine. Not sure why it isn't but perhaps that doesn't matter. It was a suggestion.

The other point was that it could replace other methods of power allocation. It seems there isn't the appetite to keep rated games on as even a playing field as possible other than perhaps using Random allocation. But that's fine by me too :)
Respect neither opinions nor beliefs; only respect the person and the right to express them.
Play by the rules but be ferocious.
User avatar
Mr.E
Premium Member
 
Posts: 160
Joined: 20 Feb 2017, 09:27
Location: Yorkshire
Class: Diplomat
Standard rating: 974
All-game rating: 974
Timezone: GMT

Re: An improved Preferences method

Postby super_dipsy » 09 Apr 2019, 14:53

Sorry Mr.E, I was a bit tired when I wrote my last post. I shouldn't have said it was silly :(

I have to say that the bits that fired me up particularly were where in your proposal for a new Prefs system you said
Mr.E wrote:At this point, nobody is penalised by when they joined the game, and everyone has as good a chance of playing the power they want as everyone else... potentially.

This implies this is not the case int he current system. But it is. To be clear to others who may be reading this, the current Prefs system is completely independent to when a player joins the game.

You also say
Mr.E wrote:Frankly, if Prefs reverts to random so quickly, it isn't really preferences.

I have to say I completely disagree. I think it depends how you choose your preferences. If everyone chooses the same three countries, or even just one, then you are right it goes to random. But people generally are smarter than that. They balance the likelihood of getting a country and their inlcination to play it.

As a proof point, I have sampled the database on preference games to compare what players specified and what they got in the game. It was very interesting. In just under half the games, only one player did not get to play a country they listed in their 3 prefs. In just over half the games, two players missed out on their prefs. There was a tiny percentage of games where 3 players missed their prefs and no game where more than 3 missed. Of players that missed their preferences, it was noticeable that somewhere around a third missed because they chose the same country as their 1st, 2nd and 3rd choice eg Russia/Russia/Russia.

Personally, I view that fact that so many games resulted in only one miss from the 3 prefs (and the rest mostly 2 misses) as far better than random allocation.

NB: But I may have the maths wrong. Maybe that is what would be xpected in a random system too?
User avatar
super_dipsy
Premium Member
 
Posts: 11847
Joined: 04 Nov 2009, 17:43
Class: Ambassador
Standard rating: 1000
All-game rating: 941
Timezone: GMT


Return to Suggestions

Who is online

Users browsing this forum: No registered users and 1 guest