Skip to content
Kinematics edited this page Jul 22, 2019 · 2 revisions

From the marker section, you can see that NetTally handles a variety of different voting methods. Each one serves a different purpose, and you should consider which one works for the type of vote you are holding.

Standard Votes

[X] A standard vote

Standard votes are used to simply count up the total number of people supporting a given option. These are used when you simply want calculate which option has the most number of supporters. This can also be used for approval voting (different from the approval vote type), where each user can vote for multiple desired options, even if only one option is being chosen in the end.

Standard votes are the default used in most cases, as they are the simplest to use and understand. There are many situations where such a voting method does not adequately capture the desires of the votingbase.

Ranked Votes

[#1] My first choice
[#2] My second choice
[#3] My third choice

Ranked voting is used to indicate a preferential order among the options presented. The overall winner is determined by the voting algorithm selected in Global Options. The default uses the Rated Instant Runoff algorithm. See the page on Ranked Voting for more details on algorithm selection.

Note that ranked votes are evaluated by task, so you can use the same numbers multiple times if you apply them to different tasks. For example:

[#1][Weapon] Sword
[#2][Weapon] Spear

[#1][Background] Squire
[#2][Background] Noble

The above would give your first and second choice for each selected task.

Scored Votes

Scored voting allows you to specify an evaluation of a vote option independently of other options (compared to ranked voting, which explicitly compares them to other options). This allows you to decide on a value for an option without having to consider an explicitly better-or-worse preference order. This also means that you are basically unlimited in the number of options you can rate, since you don't have to give each one a specified ordered number.

Further, it allows you to give a degree of preference that ranked voting doesn't allow. For exmaple:

[90%][Weapon] Sword
[87%][Weapon] Spear

[85%][Background] Squire
[25%][Background] Noble

Even though the 'ranking' of the options is the same as in the ranked voting example, with scored voting you can show the degree to which you prefer one option vs another. In this case, the two weapons are almost equal, whereas the background shows a massive preference for one over the other.

Approval/Disapproval Votes

The approval vote type is different than the approval voting method that you may see in discussions of different voting methods. This is more about allowing players to vote against an option. For example:

[+][Weapon] Sword
[+][Weapon] Spear

[+][Background] Squire
[-][Background] Noble

This is showing positive expression towards three of the options, and negative expression towards the last. In the final output, both the total number of +'s and total number of -'s each vote option got will be displayed, allowing the QM to better evaluate the overall feeling of the voterbase towards an option.

Sub-votes

Any vote type you use only applies to the first line of a vote partition. If there is no partitioning done to the vote, that means it only counts what's on the first line of the vote. If the tally partitions by block, then each individual block will adopt its own rating. For example:

[90%] Highway patrol
-[X] Down the coast
[X] Then get dinner

In this example, if there was no partitioning, the entire vote would get a rating of 90%. However if the tally was partitioned by block, then the Then get dinner line would not get an explicit score value.

What happens when voting types don't match?

Suppose you have the following set of votes:

[90%] Get dinner

[X] Skip dinner

[+] Get dinner
[-] Skip dinner

[#3] Get dinner

How would they get combined?

Well, first, votes with the same content will always be considered the 'same', regardless of the vote type specified. The vote type is specific to the user, not the vote itself.

Rank votes cannot be combined with other types of votes. If a vote has rankings, those rankings will be evaluated separately from any other vote types that get applied.

Other vote types, however, have a certain degree of equivalency. Basically, each of them can be treated as having a specific 'score' value, even for non-score votes. A standard vote is considered 100% favoring the specified option. A positive approval vote is considered to be 80%, and a negative approval vote is considered to be 20%.

Given that, any of the alternate vote types (other than ranks) can be integrated into any other vote type. Scores evaluate the score based on the above numbers. Approval/disapproval checks for above or below 50. Standard votes count anything with a value over 50 as supporting the option.

So in the above, the result would depend on what the overall vote is considered to be. If most of the voters use a particular vote type, then the vote will be evaluated using that vote type. If any of the voters use a standard vote type, then it will be evaluated as a standard vote. (These are not mutually exclusive, due to there likely being confusion over the new vote types for a while.)

What is "most"? At the moment, "most" means at least 83%, or 5 out of 6 voters.