Balancing narrative "Try/Fail" "Yes/But" "No/And" scenes with choices


I’ve spent the last few years slowly educating myself on creative writing. Of course over 99% of what you read/watch/hear about creative writing from the masters relates to conventional fiction. For example, the Writing Excuses podcast is EXCELLENT. If you have a 20-minute commute to work/school/lair, I highly recommend getting addicted to it.

Anyway, I guess I still struggle with incorporating basic narrative tools like “Try/Fail” cycles into CCH (and to a lesser extent the other project I’m working on).

I’m guessing most folks know this already (I’m embarrassed so say that I did not, as of two years ago, or at least I didn’t fully grasp the idea), but anyway, “Try/Fail” cycles are the bread and butter of fiction. The protagonist has a goal. The protagonist takes steps towards reaching the goal. The protagonist cannot succeed quickly because then the story is over. So she tries and fails, and tries again and fails again. Over and over. The number of “Try/Fail” cycles depends on the length of your story and the genre, along with your personal preferences, obviously.

So basically when applied to interactive fiction, the author would prevent the player from “wining” too often along the way so that tension can be maintained, similar to the tension in a conventional novel.

Conversely, If the player is allowed to string together win after win after win, and then “wins” the game at the end because of high stats, the story may suffer from a complete lack of tension. No tension = no real story.

Now it’s possible to style your story as a “thriller” that allows the player to win showdowns along the way, but that would still be covered by the “Yes/But” tool where the protagonist wins but is then faced with a bigger challenge. Then she wins again, but again is faced with a bigger challenge. Sort of like a “save your mom, then save the city, then save the world” sort of story. So even though the character can “win” a lot, her goal keeps moving and she’s still chasing it. So there’s still tension.

However in many stories outside the “thriller” genre, the protagonist will simply fail and fail and fail, and sometimes the failure will be accompanied by even worse circumstances. “You failed to defeat the wizard, and now he locks you in a tower.”

Okay so enough rambling…I guess my questions is:

What tools have people used to incorporate these techniques in their interactive stories in a way that still allows or gives the illusion of allowing the player to still make meaningful choices?

In CCH 1, I forced the antagonist to fail…a LOT…and it was intentional on my part, but some readers complained about being too weak or being too useless. I would like to think that I could have avoided those complaints somewhat by writing a bigger payoff at the end. However, I felt I had written myself into a corner by having the MC facing off against someone 1,000 times more powerful than the MC, and so failure at the end was the only plausible option; I guess I felt just the MC surviving and (if passing a high skill check) getting one small shot in on the bad guy would be satisfying enough.

For CCH 2, I am writing (I think) clearer Try/Fail cycles. There are definitely a ton of set backs, yet again. But I have written a more decisive victory path for the MC at the climax, so hopefully that payoff will feel satisfying to the reader. Still, I struggle with offering meaningful choices throughout the story when the MC MUST lose Battle A. And Battle B. etc etc

I offer the MC choices about relationships, and motivations, and which path they want to take to better themselves, but no matter what, each path will involve a lot of failures, setbacks, dead ends, etc. Depending on some skills checks, your MC can do poorly or “okay” in some fights, so the scenes may read differently, but losses are assured either way.

Sorry for being so wordy, but I’d love to hear some thoughts on this.


Nope. You’re not alone here :upside_down_face:

In my WIP, I think I’ll use the “Yes/But” tool where you pursue your goal (and perhaps win/fail on it, triggering “No/And”) and realize that there’s more than what you know.

The reasons behind this is that… well, I think I’m kinda hate the “Try/Fail” tool. IMO, the way “Try/Fail” works is just… umm, predictable, which is what I try to avoid on my WIP. But on the other hand, this one is easier to handle and doesn’t require a huge story plot (so probably I’ll use it only on mini-scenes/interlude-scenes).

Another reason is that, with the giant scale of my worldbuilding progress, I can do multiple “Yes/But” without worrying of running out of story ideas anytime soon, which is the downside of the “Yes/But” tool, IMO.


It seems so genre dependent to me. For example, a comic story might deal with that problem by having the main character win in a way they don’t want, or realize as they win that they don’t want to win, or fail so spectacularly that they completely destroy what they are working on, or something like that–as long as it is funny.

If you can establish a relationship with the reader that assures them that failing a test is just another name for exploring this funny path through the narrative, that creates comfort for the reader. The narrative stakes/tension is low because the aesthetic stakes have been raised–I’m wagering that I can amuse you no matter what happens, but I’m giving up a sense that terrible things can happen to you–but then, I work almost completely in various comic and pastoral genres, so that’s where my mind is.


BTW, I actually am not quite familiar with CCH, so… forgive me, ppl. :raised_hand:t4::sweat_smile:
*takes cover against the rain of tomatoes and eggs

But take a look at this vid. There’s a pretty good explanation that related to utilizing “Try/Fail” tool at it (as well as general guide at creating an interesting story plot)


@Szaal, I’ll definitely check at the video this weekend!

And yes “try/fail” can be somewhat predictable, but the antagonist can still fail in spectacular, or funny, or emotional ways. Think about The Princess Bride. The folks on Writing Excuses have analyzed that movie so many times! One of the guys on the podcast claims that he counted Montonya’s failed attempts to avenge his father, and that he fails like a dozen times before FINALLY getting his revenge, and that’s why it’s SO satisfying to watch!

I use a lot of “try/fail” but I also have at least two “yes/buts” in CCH Part 2 and I think they are both fairly exciting.

@Gower, I enjoyed MNC so much because of the prose, to be honest, I would have enjoyed it even as a conventional novel. I loved the writing style and the setting and the wacky hijinks, and frankly I don’t remember putting a lot of thought into my choices; I mostly just wanted more of what I was already getting. So I guess that’s my long way of saying you did a great job making that implicit promise of "you’ll have fun no matter, what! I promise!:

However, not everyone is skilled or experienced enough to write prose like that.


With games, this is a big issue but in gaming they usually phrase the question in terms of balance.

Due to CS stories also being “games” I tend to think in terms of: Mechanic structure such as; pipes, sinks, drains and faucets.

The right balance is an ideal, we all chase, sometimes succeeding in one area while not in others. In CS games, “stats” are the building blocks and the stat_charts are the structure. If one or more of the building blocks are off-kilter (or out of balance) then the cycles writers talk about will not succeed here.

The “opposite pairs” are a very powerful tool to use but because they are a powerful devise, the balancing on them is extremely hard to accomplish. From choosing “labels” on pairs that are considered actual opposites to determining the maths involved in setting and changing the variables - quite a few things can and sometimes do go wrong in setting up your structure with them.

Straight variables are slightly easier to balance (imo) but they are often set up as one-way structured mechanics. Your strength only goes up over time in the story, as does your intelligence, your charisma or anything of the ilk. Usually, balance is broken here by “stats inflation” and always needing to reward the reader/gamer.

Sometimes this is ok on the first publication but on sequels or prequels it sometimes forces retcons and back-peddling to happen by the author or developer.

@Eric_Moser - In your example of the final boss where failure would be the only option, if I were providing feedback on this encounter, I’d have suggested you phrase (write) the success as development for the future as the character is expected to be imported into the next title. This does not address the issue for the single title player but without further analysis and breakdown I don’t think it would feasible.

Offering the bigger pay-offs would only lead to further stat-inflation and I don’t think it would have been a long term solution.

My caveat with all this is: I am still an unpublished writer and my background is gaming.


The “fail-fail-fail-WIN” pattern works great in static fiction, but I’m not sure it works so well in an interactive environment. Most players, I think, dislike losing and want to win every step of the way. Each subquest or battle becomes, for the immediate moment, the whole quest or the whole war. At the same time, “win-win-win-WIN” only means something if there is a strong possibility of failure at every juncture.

It ends up being a question of what winning and failing actually mean for the story. Perhaps failing at a subquest still gets you the macguffin you set out for, but you lose the respect of the love interest. Or perhaps you lose the macguffin but get the chance to bond with the person who does win it. If there is only one possible outcome from a confrontation (“Your princess is in a different castle!”) then you’ve got to ask yourself what the player gets as a consequence of their actions in it. If I can only lose, and if it doesn’t matter how I play the game, then I’m going to start wondering why I’m playing at all.

(As a note, winning can have its own price as well. In my own WIP, you can gain respect with success, but there are a couple of challenges near the end that are actually easier if your have less respect: the villain’s henchmen don’t take you as seriously, or your “wounded gazelle” ploy is more effective.)


@Eiwynn, I love your analysis of opposite pairs and straight variables. It’s interesting hearing from folks who lean more toward the gaming side of things.

@Miseri, Well I guess my question would be, “Yes people (especially “gamers”) hate to lose, but doesn’t some losing NEED to happen for the story to have any real power?” I don’t mean to say that every loss has to be a total loss, so I apologize if I made it sound that way.

Yes I can see a “Try/Fail” where the hero is trying to capture the villain. The villain gets away (again!) but hero saves the innocent citizen the villain had kidnapped. I’m not saying the antagonist fails at EVERYTHING, I’m just saying using the “Try/Fail” cycle, the antagonist will fail at accomplishing the overarching objective, time and time and time again, thus pushing the narrative forward, regardless of choices. I think the details of the journey, if they are dependent on choices, can potentially be interesting enough to allow for the main checkpoints A, B, and C remaining the same no matter what, but perhaps others disagree. I’m sure many do.

And I love the way you used a “positive” stat in a negative way!


And I guess to focus things back to the original question, @Miseri does address the specific question here.

Do the demands of interactive fiction and the expectations of its readers/players make “bread and butter” conventional story structures like series of “try/fail” or “yes”/but" scenes inappropriate, or do they they still work, perhaps with tweaks?

I’m sitting here struggling to remember any CoG or HG well enough to break down its basic story structure…can anyone give an example of one that uses any of these structures?


Well, in that case, “win/fail” depends hugely on perspective. Taking a step back and looking at my own WIP from the larger perspective of what the nominal goal is, I’ve got a “fail-fail-fail-WIN” story too. But the individual subquests can still be wins on their own terms–there IS an outcome that’s clearly labelled “SUCCESS”. You’ve won the level, defeated the dragon, saved the village, and gained the respect of the citizenry … but the princess you came here to rescue is in a different castle. While it may technically be a fail, it still feels like a win (and I would count it as a win) especially if you come out of it with something that could help in the final battle.

Taking the example of a confrontation in which the villain gets away but you rescue the innocent citizen: If the stated goal of the subquest is “defeat the villain”, then that’s a fail; but if the stated goal is “protect the innocent”, then that’s a win. The trick is in shifting the player’s expectations before the confrontation.

If we’re looking at precedent … hm, I’m very fond of “The Hero of Kendrickstone”.

I would call it a “win-win-win-WIN” structure, and it’s possible here because the main villain remains a background threat through most of the story. If I remember rightly, the first quest is completely independent of the main villain plot, and it’s only on the second one that the main villain begins to move into the foreground. But on that second subquest, our focus is not on the main villain at all: our “win” condition is something else. The third subquest brings us in direct contact with the villain’s forces, but, again, our main focus on this subquest is not defeating the main villain but something else. In this way, we can still “win” each time without actually winning a battle against the main villain.


With branching interactive fiction, surely the fun is that you can make it so winning at A makes you lose at B, or that you can win at A and B but not C, or A and C but not B. Thus making sure the player can’t just win everything by having the right stats and forcing them to decide on their character’s priorities.


I agree about the writer needing to manage the reader’s expectations. Reading your posts and looking at your bio, I think you might know what you’re talking about! :slight_smile:

And I think CoG suggests weaving in enough plots/subplots so that by the middle of the story, the player is forced to focus on the player’s individual priorities, but of course there is where the “arm” ends and the “hand” begins, and that’s where this breaks with conventional fiction. We as the writer can’t write about/focus on the importance of Plot A throughout the entire story because the player may decide that resolving Plot C is more important to them. I’m doing this with my one project, but with my sequel, I need to march towards a common ending (just as I did in Part 1) because I can’t jump off from multiple different places for Part 3. I guess that’s a big part of my struggle…when writing a trilogy, Parts 1 and 2 are basically all “arm” with very little “hand.”


@Scribblesome Agree completely with this, and this is why the achievements are so great. A player can do really well on their first playthrough sticking to one strategy, but looking at the achievements will notice some interesting ones they didn’t manage to get and think “Hey, maybe I should replay with a new strategy to get this certain thing.”

I think the possibility of different ways to win, different narrative points to unlock using certain strategies, is what gives a good illusion of meaningful choice, because the player knows there was another way to do that and something they could be missing. Making them obvious, using achievements or mentioning it in narrative, is a great way to go about this. I think this is pretty similar to a Yes/But structure, but to the player it seems like a complete success either way.


One thing I’ve noticed about trying and failing in choicescript games is that in some games they end up reinforcing each other, leading to an effect where you’re either a total winner or a total loser.

The culprit here, in my view, is stat bonuses for success. If your stats keep getting better each time you succeed at a stat check, that produces a situation where you need to win early on if you want any chance to continue to win. You don’t really get a situation where you rise from defeat and learn from your mistakes… you don’t get the same kind of underdog story.

This isn’t as big of a problem if stat gains are separate from stat checks. I’ve seen this in a bunch of CoGs, where earlier choices will set your stats, and later choices will determine success/failure based on those… it has less of a stat inflation result.

I did have a thought about an unconventional possibility, so I’d be curious what people think of this… in regards to the “learn from your mistakes” idea… what if, when you use a stat and fail, that raises the stat, rather than when you use the stat and win. You’d get kind of a trade off between narrative victory and learning… and that might help create a situation where you can get more of a win-fail cycle, more of an underdog achievement story, and prevent a cycle of win-leads-to-win. (I’d be curious if this would work in practice :thinking:)

In this analogy, what would be what? :thinking:


With a static narrative, it’s difficult to tell exactly how much effort the protagonists are expending to stay ahead of the antagonists. Simply writing “Wow! That was a tough fight!” isn’t really enough to convince most readers the fight was actually tough, or the protagonists actually had a chance of failure. Thus, a mixed cycle of failures and successes is generally a good idea to build tension and reader engagement.

With a game, the players will know exactly how much effort they’re expending at any given time, even if they’re unsure how hard their protagonists are working. They may even have a fairly reasonable grasp on how much they can relax before slipping into failure. This can build a level of tension and player involvement into the game independent of the protagonists’ win/loss records.

I'm going to try to make a silly example. I'm not actually sure it will help me explain, so I'm tucking it in a details box.

Let’s say I have two games where the protagonists are sneaking through enemy fortifications, attempting to eliminate antagonists while remaining undetected.

Let’s say that in one game, success is entirely stat based, and my character has exceptional Stealth, Melee Combat, and Sneak Attack skills. All I have to do, as a player, is guess which options in the choices amount to “Sneak up and stab somebody.” Will that game be able to maintain tension and engagement without a “try/fail” cycle or “yes; but…” results to my choices?

Let’s say that in the other game, my character still has exceptional Stealth, Melee Combat, and Sneak Attack skills. However, let’s also say this game recognizes my character will be a blade in the dark, and offers me multiple blade in the dark options, while hiding spellcaster, social manipulator, and armored juggernaut choices. Now I can’t simply choose the option that panders to my stat spread—all the options do! I have to read the narrative carefully and choose based on the environment depicted and antagonist actions described. Will this game still require a “try/fail” cycle or “yes, but…” results?

I suspect most multiple-choice text games can get away with a notably lower level of “try/fail” or “yes; but…” than a traditional static narrative. However, I also believe they’ll need a notably higher level of such results than, say, a sports game, tower defense game, or first-person shooting game.

Looking back to Mecha Ace, (mostly because I’ve played it fairly heavily,) many of the choices I found (and continue to find) most tense were not actually in combat scenes. They were choices like “Do I defend the fleeing civilians or the military fleet full of my allies?” before the combat opened. Choices that are full of the “yes; but…” dilemmas. “Yes, you saved all the civilians, and pushed back the imperials, but your allies took heavy losses.” I’m fairly certain there were some “try/fail” cycles as well, but they weren’t memorable.

As @Scribblesome said; choosing between “winning at A and B, but not C” and “winning at A and C, but not B.”


These responses are so thoughtful.

I wish we had more “craft of writing” threads like this…maybe I just miss them when they pop out.


FailBetter games (who made Fallen London and Sunless Sea) work kind of like this, with stats going up on failures as well as successes. Stat checks in those games work on random number rolls, with better stats giving higher probabilities, so while it’s possible to just get lucky, usually passing a hard check will take a lot of failing first.


CoG suggests that you DO reward the player with a stat bonus when using that stat, even if they fail the stat check.

For example…

Player chooses to pick the lock!

Success requires lock picking of 40%, and the player only has lock picking of 30%.

The player fails the check, but the player’s lock picking stat goes up anyway.


Re: raising stats if you fail: I’ve been wondering too! I’d guess that’s something for the CoG team and their expectations … I know they’re against stat checks that reward the player with an increase in the same stat being checked–it always has to be a different stat, if there’s a stat reward. Stat increases for failure could also solve the “Is this choice going to increase Stat X or is it going to test Stat X?” problem.

(Aside, to @Eric_Moser: I have a bio? Huh, so I do. I’d forgotten … I need to update this.)


This is going off on a side-bar but with Eric’s indulgence I will expand.

Variables are the mechanical building block but it is the scripting that you do that builds the structures in your mechanics.

A pipe, which takes the narrative from one place to the next, is built using different commands such as: *if, *elseif and *else which have stat-checks involved as well as *choice, fake_choice and things like multiplechoice and nested *choice

A sink is an “encounter” scene or a “resolution” structure. Again, it is built using your scripting code. This is usually in the “body” of the chapters in normal fiction, with IF it can be located almost anywhere, which can cause problems for an author if the rest of the structure is not firmly connected to the “sink”

Drains are the “fail”, “but” and “and” of the writing devises Eric mentions. This is where stat building blocks (in theory) should be reduced or even taken out. This is where a lot of new developers fail at balancing. They never take out, only add. Once again, this is accomplished by script/code.

Faucets are the places where more of your building blocks are added to the mix… again built with scripting commands.

I too wish this.

All I am going to add to this is: beware of stat inflation and the need to keep everything going into future titles (assuming the project is not a one-of.)