Opposite pair variables vs. single percent variables

Continuing the discussion from COMING THURSDAY: "Cannonfire Concerto" - New Author Interview & TRAILER:
@idonotlikeusernames @poison_mara

Branching off from the discussion in other thread, just interested in how players feel about opposite stats vs. single stats, which fit play better as a gamer or / and as a role-player. It might depend on variable, but then what factors within variables determine whether better as opposite or single? : )

  • Opposite stats
  • Single stats
  • In-between e.g. not opposite stats for skills (please explain specifics below)

0 voters

1 Like

Each author needs to design a mechanics structure to ft their needs. Having opposite pairs or unidirectional stats is neither good or bad.

Mechanics serve to connect the reader to the game and if there is a dissconnect then there is a problem in the design.


Opposite pairs only work with stats that relate inversely. When you get stats that don’t always correlate – such as work/happiness that’s where you get into trouble.

I am happier as a person the more I do productive work, and so when I see paired stats that pit those two ideas against each other, I don’t relate. I’ve seen several games that fall into that trap.

In my view opposed pairs work best with “side A/side B” situations where you can’t be on both from a plot perspective. That flags a reader that they need to pick a side, and is a quick way of showing the value conflicts in the game.

Heroes Rise also did opposed stats very well. Lawful/Lawlessness; offensive/defensive; solo/team player – those are excellent, if cliche, examples of stats that make sense as paired.

The key is really “can you be both of these things at once?” If you can, don’t make them an opposed pair.


I think my feelings on the matter are probably mostly in agreement with a lot of those I’ve seen on the Cannonfire and earlier Empyrean threads. Basically, an opposed pair of stats is fine for a personality or morality pair of stats, as long as those stats are truly opposed (at least for the purposes of the game).

A second point is that I feel such stats should generally be used for flavour text, not for success vs failure. Example: a “selfish” character might feel contempt for the struggles of others, but they should be just as capable of helping people as a “selfless” character. Sure, a selfish character would probably not help for nothing, but maybe they believe that they’ll be rewarded; likewise a selfless character may choose not to help for completely unselfish reasons. The only times it should be used as a success vs failure would either relate to mind reading (or similar), or potentially due to NPC reactions if the MC is famed for their selfishness/selflessness.

Finally, there should also not just be a cut-off at 50, but rather a band of grey (maybe 40–60?), over which the character is balanced between the two (again, almost purely for flavour text purposes).

That said, I feel that there can be exceptions to this, for example in the WIP Through Broken Lenses by @Interestedparty, in which (for example), “combat” is paired off against “friendliness”. This works because of the style of the game, which is all about trying to find a balance between being an amoral killer with no friends and a social butterfly who can’t protect the friends they now have.


I find opposed pairs much harder to write for. For a set of unipolar stats, if I want to award or deduct points for a choice, I can simply say %+10 good_breath or %-10 stir_fry_ability and know that the player is gaining or losing, assuming that it’s always better to have the best breath possible or be the best stir fryer possible.

But for opposed pairs, if I have sassy/sweet as a pair, and I have a choice that gives %+10 sassy, that may a reward or it may be a penalty, depending on what the player is trying to build, stat-wise. So I find it harder to figure out how to balance.


Here I think I would have to disagree: I don’t think opposed stats should be used wholly for flavor text, because they can be very useful tools for tests.

In my current game, I have a single opposed stat, which is abrasive/soothing. That is of course a personality indicator, but it is also something that might be tested, and frequently–are you able to soothe someone/annoy someone is too juicy to not test.


First I love your writing skills. I loved so much Mid summer. And Empyrean writing. However, Empyrean is the most extensive example of how do wrong an opposite stats system. And why It’s dangerous trying out an only opposite stats game.

As a rp is impossible play your game . Except you have a very unchanged perspective of a character that just happens to be metagaming. You have to be a brute tech or a emotional art lover. And cunning with potato skills on people (that has no sense cunning people normally are good with people. They needed that to lie) or a Straight forward zero political idea leader…

If you try to do something in between more normal, you pass from a brute lockpiker that was always great in locks. To a emotional artist who never tried a lock in few chapters.
Your new game seem promising please don’t make same consistency errors …If you need help you could contact me anytime.


Opposed pair or single stats could both be great as long as they are used well. Stats should let the player role play as a person who’s personality is not completely one thing, and no one should be forced to choose a certain option because they think their stats would make them fail when their character should succeed.

Like in a lier versus truthful opposed stat, a person may tell the truth a lot, but come across a specific situation where it would be useful to lie. A stat check would make them fail, but realistically a person who is usually truthful can lie occasionally.

Opposed srats are better to check how the npcs view the MC. Like the MC could be more trustworthy of a high honesty MC, or the MC could think the MC is evil because of npc’s definition of evil. Like a rebel not trusting someone with a low rebellion score.


So if for example, I’d based stats on two archtypes that play opposing roles, because it fits the game style + narrative, could they be opposed? Can you be both e.g. villain + hero (that’s not my stats just simplest examaple ever)? Yes, but then maybe I place more importance on the minority stat’s percentage still present… Would problems be solved if I left options to be grey as well as black and white?

62% / 37% makes it clear that at the moment, almost a third of people in forum prefer single stats. That’s quite a lot of people disliking opposites, and from comments, it’s maybe because some people think opposed stats add a more natural human dimension of grey duality, but others think opposites limit it. It’d be a lot easier to decide if people’s preferences were based on different aims, instead it feels like ppl think different methods will achieve the aim of true realistic narrative + immersion better. So prehaps what matters is the realism and intricacy behind the stats?

1 Like

Yes, Opposite stats could working but it is very difficult and expensive in time versus single stats that mechanic talking offer way more versatility. Opposite stats are a single variable in game not two so you cannot separate both points of the spectrum . If good and evil were separated stats you could code if evil is plus than good or if each amount of evil and x amount of good happen you got x grey actions or flavor text If it is opposite you can’t neither of that you have only a stat moral .

1 Like

That’s true, but would you object if you had a high truthteller-to-liar ratio, and you failed the stat check because you just aren’t used to lying, and you are bad at it? Or is it that you want to at least have the option to try to lie (even if you fail)?


Yes, yes role-playing wise and option to lie even if will fail or is random chance success could add so much deeper into the game . It adds a touch of reality.


Also, it is possible to check if an opposed stat is, say, between 35 and 65 and have a positive result for characters who are in the grey area. Maybe characters aren’t quite sure if Character A is lying or not because the Lies/Truth stat tends to hover around the middle for that character.


I feel like skill abilities should be single variables. Like the more a person lies doesn’t make them less able to tell the truth, or the more a person exercises doesn’t suddenly make them more stupid and forget how to read.

The frequency of lies to truth could be an opposed pair, but not skill. If the frequency is significantly skewed in one direction like 80 percent or more for one variable, then that could reflect skill. Like I learned calculus and high maths in school years ago, but I don’t remember the skill now because I never use it in my day to day life. But, this is a grey area, because there may be some skills that are never forgotten like riding a bike.

If my lie skill is low, I should fail, but I would like the option to try.

1 Like

I could imagine a world in which one could account for a tradeoff of the kind you describe. If you focus on training for your track meet, you might not be as sharp for the test. And I think the lie/truth thing is even more imaginable. If one tells the truth almost all the time, one might be really bad at lying. Doesn’t mean you can’t try it, but I think it could make sense to have that be an opposed skill.

1 Like

I don’t disagree with these examples, but the problem is extent of the change. Practicing for the track meet shouldn’t make you stupider by an equal amount to what you gain athletically. So the example isn’t bad, but it doesn’t make a case for opposed stats, but rather for a small subtraction from a single stat or perhaps a modifier to a test_difficulty variable or something.


But, no one would believe a compulsive lier if they knew the person. A compulsive lier lies about everything even if you saw them in the act. If a person believes the lies of someone they know to be a lier or saw them do the thing they lied about, that person is incredibly stupid and unrealistic. Like how many times did Shaggy have to say it wasn’t him before you believes him in the song “It wasn’t me” (song might be a little NSFW).

Spending more time at the track could mean you spent less time studying for a test, but I agree with @hallofmirrors unless you stayed up all night at the track and fell asleep or overslept for the test.


It’s important to consider the implications of having opposed pair variables when deciding whether to use them. When you use an opposed pair for something, you are saying that the two stats are linked such that increasing one neccessitates an equal decrease in the other. So something like Sabres of Infinity’s Cynicism/Idealism makes sense since you can’t be both cynical and idealistic and when you become more cynical, you become less idealistic. And if you start cynical and end up idealistic, you haven’t forgotten how to do cynical things, you simply have a different perspective. But when you make two abilities opposed, that makes things very odd because it’s very difficult to find two abilities that work this way, so you create world where someone who both studies hard and exercises can’t do either well.

Edit: And what is important about, for example, Cynicism vs Idealism is that it ends up producing a good representation for your actions–someone who makes many idealistic choices will be more idealistic and someone who makes both with roughly equal frequency will have more centered values. This is good because most characters are going to be more or less in the middle, which reflects the real world–most actual people are not at either extreme. But when game mechanics and real world examples come into conflict, the stats don’t provide a realistic representation, which frustrates players and makes for nonintuitive gameplay.


Exactly that’s the reason opposite stats should be used by morality or personality traits well chosen not for skills or abilities. Like sport vs study Wtf ? there are several sport players who lately work in scientific research or lawyer and entertainment. Sports is a good way intelligent people get a scholarship. However thousands of games believe that are opposite stats.


First of all, this is a great conversation.

Second, let me ponder aloud to add to @poison_mara’s categories of moral and personality exceptions. For the sake of argument, my responsibilities break down into two things: scholarship and teaching. I have limited time, so I have opportunity cost for everything I chose to do. I also have a skill test coming up, let’s say (promotion, or tenure review, or whatever).

So I could think of my time as the whole pie, and every time I gain in scholarship, I lose in teaching, as a percentage of the whole. I wonder what the best way to represent that in stats would be. That’s not a moral or personality trait–I guess that’s a resource: time. If resources are zero-sum, that’s a good reason for an opposed pair.

But here’s the tricky part: let’s say I want to then see if I write a great article and I want to test scholarship. Would you like to see another stat that measures scholarship? This is hard.