Opinions about Stats (Skills, Relationships and Personality)

Hey. My cute little robot singing “Maps” by the Yeah Yeah Yeahs IS cool content, okay??? (I mean, have you seen that video?)

I do like surprising people with paths they didn’t think were possible, too. But, I think it’s also important to write with the understanding that a LOT of people will never play the game more than once, and some won’t even finish the first time. I can’t tell you how many people have told me, “Yeah, I started Choice of Robots” and it comes out that they basically only finished the first chapter. Which is too bad, because it gets better. In fact, I actually cut some early life stuff in the first chapter because a playtester didn’t even make it past the first scene when it started out with a five-year-old MC. “That’s cool, you’re like a five-year-old in this game, good luck with that” is a paraphrase of what he said, and he found something else to do at that party. And he was a pretty smart guy.

So to get back to the issue of stats: that’s what we’re up against. Not the perfect reader who plays ten times and catches every subtle allusion - they’re actually easy to please, because you like the same things - but the guy who on a scene by scene basis is deciding whether to actually watch Netflix instead. Smart guy, and he’ll like your game and give it five stars on Google Play if he really gets into it. But you’ve got to fight for his attention, and stats are a way of putting more of the game up front.


I like to think of CoG more as an interactive story than as a game. There are two main types of readers- those who only play once and those that like replayability. An author wants to try to satisfy both groups.

Stats can be a good way to add replayability without writing completely different stories. If the MC has to be accused of a crime and exiled for the story to work, allowing NPCs to look apologetic and secretly voice dissent for a “good” MC but be too scared of the strong government to do anything ,would better the interactivity without a whole lot of work. It might not make sense for a NPC that the MC has been cruel towards to want to help the MC, however adding an extra sentence about the NPC only helping because they believe the MC is innocent and because of their view of justice and not because of a good relationship, adds some variability without a lot of work.

Stats can be done poorly and lock the MC into one specific play style when sometimes it’s better to use finesse than brawn or you’re scared to take an action you want because you think it’ll end poorly. Some people like seeing stay changes because it immediately shows that their choices have some impact (even if the stats are never really used) and some feel stats are restrictive especially if there is no clear connection between how choices impact the stats and what the stats mean. I would not immediately not play a game just because it had or didn’t have stats, but I do feel games with less skill stats are better.


I’ve purposefully stayed out of this thread for a bit to get a larger sample of what others think before writing my perspective.

I’m going to start with the following:

My background is gaming and the gaming industry. When I see authors approach the CS script as just a glorified text editor I cringe inside. The stats and the means to display those are mechanics and should be viewed as part of the structural integrity of your story. Most authors do not grasp what this means. Stats are not just to add interactivity to your story and used in a superficial manner; if you do this, you will often lose a major part of your audience.

Your audience is a full-spectrum and not just ven-groupings of two types. There are as many motivations to playing a game (and CS games should be acknowledged as such) as there are players that actually do play the game. They may share some commonalities but I guarantee you that their differences outweigh the commonalities.

Because differences outweigh the commonalities, your job as a developer of games is to find as many points of nexus as you can. A mechanics structure (of which stats is an important building block of) is one such nexus point creator. By signalling that a relationship is strong, you form a point in your game where all the different readers and their motivations can collectively relate to common ground. This is why relationship wars and shipping fan-fiction/fanworks thrive so much.

Stats also are a visual way to relate information to those that learn from visual cues more readily than from written cues. This is more necessary with CS games because you want to reach all of your audience as much and as clear as you can. Most games are visual, so cuing a horror-filled face at your lack of humanity is easier via art then it is for us when dealing with the more visual of our audience. Stats and their displays are where we can visually relate our written word to those who are not able to get the concept completely through written prose.

A well executed mechanical structure is necessary with CS games because of the inherit purpose of the CS script. Stats showing visually what we try to write in prose, is a way to relate the mechanics of the game to our audience and a way to explain what is happening under the hood of our game.

Others have talked about sign-posting content and signalling important mechanics. @kgold has wonderfully laid out various reasons why this is good and I agree with him. I also agree with @Gower that each path in a game is its own reward (at least theoretically) but practical experience over the last decade has taught me, painfully something more. To be successful, a game’s utility needs to be constant. Stats and their display provide the tool which often even outs the utility of a game to a constant from beginning to end.

Stats will tie the enjoyment received over the content of any one branching and it will provide a way for your audience to translate their “happiness” received at any one point in that branching. Mechanics tie the entire game together and they promote all the various purposes your audience may wish to accomplish. Romance, conflict, resolution, learning, and whatever else you can think of is facilitated by mechanics - stats are just one tool of the mechanic toolbox to use.

Stats a mechanics are often the easiest to use because most gamers have been exposed to them, their entire gaming life and as such they require more skill in deployment to get the same “bang” that our other tools need. “Everyone” knows what “strength”, “charisma”, “intelligence” and “agility” mean - or so they think. Which is why when they are presented in a structure (such as opposite pairs) that confuses, frustrates or mystifies our audience they will complain.

This is a major weakness in many author’s ability to write CS games and it is often a reason many gamers score their titles low.


I have a general question for writers.
I don’t doubt that creating a game requires a lot of careful planification. So, how was the process that make you choose what stats you wanted to include in your games? I feel that hearing about the experiences of others during the development of their games can provide a better insight in these subject and help new writers.


For me, it’s all about the theme I want to explore. In Midsummer, the theme was about the pull between responsibility, silliness, and freedom, so the three major derived stats all linked up to that.

But on a broader level, since Midsummer was comic, I didn’t have a “combat” stat, but instead a “courage” stat which could be used to fight, but also for facing challenges head on. That makes a more flexible stat. You can think of more ways to use “courage” than “fight.”

Even better, that lets you look at your stats and think of your character’s personality, not just what you can do.

Ideally, you can look at stats and know what sort of game it will be–its genre, and the sorts of themes it will deal with.


I tried to pick the stats with both theme and gameplay in mind.

A key theme of the forthcoming Choice of Rebels is that when you’re in revolt against e.g. an evil fantasy Empire, it’s often non-obvious where to stop rebelling, or whether you even can. Where does a corrupt social order retain some value, and how much can you destroy before you lose the ability to build back? What if anything does your rebel leader MC value as a source of social order?

The game’s three opposed stats are all centered around major potential sources of social order – nationalism (or cosmopolitanism), religiosity (or skepticism), and ruthlessness (or compassion). These values affect every stage of the rebellion, not just the rebuilding (which we won’t get to for a few games yet). And, importantly, I find each one fun to write. There are interesting trade-offs and dilemmas and colorful outworkings for each side of the stat.

There’s also an Anarchy stat which is visible as a number on the stat screen (zero, at the beginning). Other important variables like notoriety and morale aren’t ever shown to the player as numbers; they’re periodically suggested in the text, so the reader gets a sense of whether morale is e.g. strong or dismal, but not whether it’s 2 points away from some critical threshold. That’s because the artificiality of having too many stats in precise number form (rather than described in text) can be immersion-weakening for me. But the level of Anarchy you create is so vital to the theme that I want it visible to the player throughout the game.

I also included three skill stats – combat, intellect, and charisma – because I thought they could lead to fun variability in the gameplay. There are distinctive ways of rebelling as a great general, an effective user of the world’s best thinking & technology (in this case, magic), or a charismatic leader; writing three different specialisms feels like it should be manageable. And as Gower suggests, the choice of stats here also signposts that if you like fantasy RPGs, you should give this a try. (Nothing like calling one of your stats “charisma” to give the game away.)


Agree with both @Gower and @Havenstone on building around a theme. We wanted our stats in So, You’re Possessed! to be more literary in nature and revolve around character development, as opposed to needing to build up super powers, magic, or strength for battle. Not that there is anything wrong with that kind of game, it’s just a different flavor. With a story and humor driven game, even though there are fight scenes, we figured any stats we use will need to complement that in a way that emphasized personality and worldview over combat mechanics. And as mentioned above, being more assertive might also give the MC more to say and do if they are outgoing, and naming that stat “Confidence” as opposed to “Courage” or “Will” helps further designate it as an urban fantasy as opposed to a DnD style RPG.

Our MC’s choices and resulting stats do have a more subtle impact on the reader’s opportunities and the tone of the story in ways that might not be immediately obvious within the text (i.e. characters who choose more hostile/aggressive options may be met with more hostility/aggression than those who don’t).


For designing the stats for Choice of Robots, I started with the idea that I wanted extremely different climax chapters, and I wanted the stats to decide which one the player got. So I thought about what different kinds of stories you could tell about robots - robot revolution, robot utopia, robot love story, evil genius robot army - and made a stat that corresponded to each one - Autonomy, Grace, Empathy, Military. I’d have edited them if that didn’t provide enough options for handling situations, but they seemed okay for that, and I really liked the idea that the moment-to-moment decisions were also determining the genre of story you would have in the end.

I also thought it would be interesting to make the robot have all the stats, mirroring an MC viewpoint that emphasized the robot over himself or herself. So the protagonist just got one stat, Humanity, that would be silently drained even as the robot’s improvements got shout-outs in the text. This was very much not letting the player be exactly who they wanted to be, but I thought it was more interesting at the time to put the player into the skin of a person who had definite traits.

For Choice of Alexandria, I don’t think the process worked as well. I started with the idea of having Ptolemy IV’s pettiness be one stat and how much he liked you another, and then said, okay, that doesn’t really give the MC ways of dealing with different situations, so how can an ancient Greek polymath adventurously handle things? There are three skill stats because CoG requires at least three options in every choice. Then they asked me to add some opposed personality stats, which I did, but I didn’t have a great idea of what to do with them, and I think it shows in the game.

For Choice of Magics, the skill stats are based around five fantasy versions of technology that could go wrong - Negation for nuclear power, Automation for exactly that, Glamor for mass media/pop culture, Divination for the Internet, and Vivomancy for bioengineering. The theme’s deciding whether people can be trusted with these double-edged swords, and the player silently acquires drawbacks in the world even as the skills are called out, much like Choice of Robots. I also wanted to give the player more choice in their POV than in my previous games, so the Empathy/Calculation, Humor/Solemnity, and Optimism/Pessimism axes significantly change the narration in places. The initial plan is to again do five different climax chapters keyed to the different magics, but lately I’ve been thinking I might want to tie the climaxes more closely to character relationships instead. After all, fantasy’s not science fiction, and it’s less idea-driven and more relationship-driven overall. It might be interesting to drive the climax chapter with a variety of factors, is my thought. We’ll see, since I’m not there yet.


First off, I love this thread since (like I’ve seen others say for themselves) it’s something that I’ve thought about a lot.

Second, my two cents is thrown into the pile of “what are you trying to do with the game.”

Though it’s not seen immediately (due to how little (relatively) I have for a demo so far), I use my stats for dual purposes. Number one: to highlight the fact that it’s about people literally winding up inside a video game. And number two: to have fun with people who love to flip to the stats page with every choice (a habit I have).

For the first, I’ve limited the stats to various combat and utility skills (like how well you can aim an arrow or brew a potion) as well as showing some relationship stats. (I say some, there’s actually always at least nine.) But I’ve included these because the player does have a very real connection to this game-world. I plan to highlight this by having the player discover later on that the others they’re with aren’t as attuned to it as they are. The actual person behind the phone or computer screen can flip back and forth between the game and the stats screen, and in-game, this basically translates to the player thinking harder than usual. The other characters, like real people, can only hazard a guess as to how “good” they are at something, based on how often they succeed in that area. So if the player mentions something regarding stats, this cues the other characters into how they’re missing something that the player is obviously getting.

That was a bit of a ramble, but what I intend to get at is the stats are there to guide the player into certain paths as a fallback. That said, I’m consciously avoiding stats like: good, evil, generous, or even strong because I feel those things to be limiting and subjective. Whenever I see a stat in a game titled “Resolve/Willpower” I immediately get a sour taste in my mouth because that signals to me something that’s going to have to be invested in if I want to play my character as someone who doesn’t give up (when it’s reasonable, at the very least). For me, I’m playing this person because they’re the “hero” of the story who is willing to go the extra mile to get X done. Also, I dislike stats like charisma because I think you should have to work out a discussion rather than throw it through a decoder ring. I plan to have stuff like getting say, a foreign king as an ally because of how you’ve acted around them, how you’re regarded by the people with you, the things you’ve done, and the things you can pick up about the king (such as noticing they value a strong, common faith in their nation over trading with the outside world). If you can “read” a person and what they want, then you have a better chance at getting what you want. Maybe you snooped around and found out the reason some noble is holding back their army is because they’re rightfully paranoid that someone else is aiming at them. Working out a situation like that is a lot more organic to me than just, “Look at this smile. C’mon. How can you say no?”

To get to the second point (with having fun with page-flippers like me), there’s always one set of stats the player can see. The “Relationships” stats. And real quick on that: I plan to have variable relationships where it’s case-by-case rather than “I like you because you always agree with me.” For those, it’ll depend on the things the two of you have done, what they’ve seen or heard of you, as well as the character themselves (some people are just more loosey-goosey than others). But with messing with some types of players, there’s one character that is introduced and then removed on the stats screen in the same chapter, simply because I find that funny.

Even with all these visible stat considerations, I also have plenty of behind-the-scenes stats being counted for different reasons. One tracks how much the player is “settling in” into a character for a typical Fantasy-RPG versus an average person, one tracks how often the player flirts with a specific person, another tracks if (barring behaving and thinking alike) someone else at least respects you enough to hear you out should you try to work with them for some goal. But I only show the nine other main characters in the ‘Relationships’ tab because the player has a literal connection to them that isn’t there for some other NPC. The player can still be chums with someone else, and even follow down a path of romance with them, but there is no explicit number attached to them on the stats screen because the same connection isn’t there.

TL, DR: I agree with people saying “it depends.” Apologies for ramblage, but this is basically a dump of a lot of the back-and-forth I’ve had for this single topic.


Oh, I could talk about stats alot, since I both hate them and love them, but I think many have already said most of what I think, I would like to add something about the relantionship stats and that is sometimes they are necessary.

Take my most loathed game ever, DAI. It had essentatially to set of companions: followers and advisors, followers had approval stats, but advisors didn’t - the result was that advisors could not really react to you in the same way followers did and it was really obvious because followers did have approval´ which meant they had a high and a low extreme.

One of the advisor, Cullen, should really not be as agreeable to my character as he was since I did pretty much everything he did not want to do - yet the game never reflected that in his opinion on me, on the opposite end I could not really express dislike for him with my character either since the game just didn’t allow for it - it had no way of tracking the relationship so it defaulted to a net positive relationship even if that relationship wasn’t one I wanted.

So, yeah, relationship stats simplify human interactions a lot, but at least it allows for the game to track the relationship - the alternative is that the game decides the nature of the relationship for you.


Tell Tale Games which makes IF such as The Walking Dead Game doesn’t have visible stats (well, at least the titles I’ve seen. I haven’t tried all of the titles) except for the “they will remember that” feature. The Walking Dead had plenty of feels without visible stats. :sob: Lee and Kenny.

Jeff Dunham is a comedy ventriloquist/ puppeteer/ guy who plays with dolls. He once said that he wants people to think of his act as a great comedy show instead of just a puppet act.

The Kite Runner book was fiction, but the characters seemed so realistic that I had to recheck to make sure the book was not an autobiography.

It would be good if someone’s work could be thought of as more than just a story/game.

Airis (suicide, PTSD, no good endings? trigger warnings) is an otome type game where the Airis characters seem more realistic (ish) than the earth-Airis characters who focused on stats.

I don’t hate stats, but sometimes the game is so focused on stats that it hurts immersion and roleplaying. I want to choose to do something because that’s what my character would do and not just to pass a skill check. It’s especially bad when a game forces my MC to act a certain way because of previously chosen actions (like in Airis). There is no possibility for dynamic characters. Stats should be more impactful of NPC decisions and events caused by the MC. I usually have more defined playable characters in sandboxes than in mmorpgs where my main objective is to grind.

Having only one or a few good ends with mostly bad ends is not fun. Having classes like fighter and mage that don’t change the overall story isn’t something that would make me want to replay. I more prefer different paths like a light and dark route and the only reason you can’t accomplish everything in one play through is because either you don’t want to or it wouldn’t make sense to do so (I get it if I can’t keep everyone alive or date someone if they’re strongly against my morals or wouldn’t stop trying to defeat me until I’m dead or they are dead).

Also, the strength of the narrative should be good enough to signal different routes. Like in the From Ashes We Rise WIP skills where the lab is investigated and evidence of your power use is seen which could be troublesome for someone with nonnatural powers who doesn’t want to be found out. People expect at least some railroading, so it can be hard to tell when a choice will actually impact something.

Numeric stats can also be a little confusing. For health, I know 100 is the best/ full heath and zero is dead/worst and anything in the middle is in the middle. For relationships, low could be hatred or maybe they just don’t know MC well and the relationship has to be built up. Textual stats like in Choice of Dragon are clear about the middle, but not about the best or worst stats.

1 Like

My main consideration was thinking about how to offer different expressions of character in the story. So the skills give different ways of solving problems, and the personality stats offer the player’s character different attitudes towards the world. The skills then were chosen based on what was interesting and made sense for the story. The personality stats were tied in to the main themes of the work, what the bulk of the choices were really about. Some stats were then picked as interesting but sometimes conflicting goals for the player to have (money, fame etc.).


A Midsummer Night’s Choice is an example of where I do like and appreciate stats.

I noticed that if I select “I change the subject by trying to engage him in stimulating conversation.” when you first encounter Father, that prior choices achieve different results. I can pick rapier for the painting and jumping across the balcony with Father being a great art patron, but I’ll give a completely different response here just by reading a book without Prenzie then if I had gone spelunking with or without Prenzie thus impacting how intelligent of a response I give and therefore how furious Father is when I say that I like the poems that rhyme. I also liked that despite whatever my character class stat is, that I could freely make choices that would either not boost it or even diminish it without hurting my opportunity to achieve a good ending.

I also liked that I could rip up all of Morgan’s love letters and still get his/her best ending because I should be able to do things that will not please them and still get their good ending because I maintained an overall happiness level with them. I loved the married as friends (with the person Father intends) for the good of the kingdom ending because good endings don’t all have to be romantic even in a romance CYOA.

Things like the above make me want to keep replaying so that I can see all the endings (even the bad ones) and collect all the achievements.

Where I do not like stats is when they are arbitrary to the point of having no purpose and really do not need to be there. It would be better if the author didn’t have stats at all in these cases. Creatures Such as We, for example, has no stats and no achievements yet was quite satisfying.

I hate stats even more when their exist solely to punish the player for making “wrong” choices. I should not have to achieve a 100% score in whatever my class stat is in order to get a good ending. I shouldn’t have to boost a specific stat regardless of what my class is in order to have a good outcome at a specific plot point. I also hate achievements that are frustratingly difficult to achieve because of stats (such as the Lawful Good achievement in The Hero of Kendrickstone.) In these cases, it isn’t about having stats above a certain level to be able to win with a certain choice but being railroaded down one specific path choice for the entire game.


(I was certain there’d been a thread for this. But apparently there isn’t. Huh)

It goes without saying that CoG/HG games have brought forth a massive amount of stats and ways to use them.

But what do stats mean to you?
What do you expect a type of stat influences?
When you see e.g. the relationship stat, do you expect it to just keep track of what a npc thinks about you? What you think about them? Or do you expect the npc’s dialogue and behavior to change? Unlock a new story path?

And what would you advice authors against when it comes to what to use a certain stattype for?

What do stats (types) mean to you when you see them on the stats screen?

1 Like

Well, i’v seen stats used both as way to ‘stats check’ or to define a character’s skillset like in many stats management game (like that mercenary WIP) but i also saw them used to determine the character’s personality (like in samurai of Hyuga) and since it’s the two way to use stats i can remember on the top of my head, i’d say it’s pretty much the two thing i think it will tell about my character

what to avoid : being unclear on what stats does what and what stats is needed for what


I agree with @Balrog_Demorgothe here when it comes to stats about the MCs skills or personality so I’ll only talk about relationship stats and why I think it’s often not handled well. Often when my MC interacts with an RO I have to pick the options which aren’t exactly what the character I’m roleplaying would say or do at that situation but I have to pick the choice which raises the relationship stat with the RO. An example would be when I would like to take things slowly bc either I would like to get to know the RO better before jumping to things like kissing and stuff or simply just bc MC is shy.


Actually there has been a thread about this

1 Like

Thanks <3 didnt find that one.
@moderators could you merge?

Personally I’m not a fan of stats. I approach IF as a storytelling experience, not a “game”. I’m not trying to win, but rather get lost in a narrative that I have some measure of control over (no matter how illusionary).

Opposed stats are a big pet peeve of mine, simply because I’ve had the experience of my character turning out very differently than I intended because doing what seemed logical to me to raise X stat turns out it seemed logical to the author to change something else.

In the story I’m currently working on I’ve somewhat hesitantly added four visible character stats to represent the MC’s story-usable skills. However, I have only one true “fail state” thus far that can cause you to miss a piece of evidence if you insist on using a stat check. In that case I have also added a riskier option that doesn’t involve the visible stats–basically you bet on your cleverness/deductive skills vs doing something you know is going to hurt but will most likely get results. Early days still, so even that is subject to change (game balance is WAY harder than producing text, imo).

I am absolutely 100% committed to avoiding the one-skill trap, where players feel forced to solve every problem using the same stat because that’s the that has increased the most. I won’t let you succeed brilliantly every time, but I won’t punish you for stepping outside your character’s comfort zone either.


Stats, though, is a crucial tool that allows for a really branchy narrative experience that is also able to be written by a person with finite time. Definitely there is great branchy IL that doesn’t use stats, but using stats with small incremental changes over a great number of decision points allows for the delayed branching that makes IL feel impactful.

I think the key point is that “stats” can encompass classic D&D strength and dexterity but also far more subtle things like “duplicity_sneak_into_cocktail_party” or “outrageous_accent.”