Word Counts, Percentages, and Stuff! (Patreon)

Helloooo everyone!

Inspired by a voice chat yesterday with a friend, I have decided to ramble at you about one particular part of the writing and planning process when it comes to visual novels - and Gilded Shadows in particularly.

I could probably do an entire TEDtalk on word count  - and I actually have considered it because it's such a weird point of contention in the indie dev world. 

I know it seems like it shouldn't be, but a lot of times when a newer developer asks about target word counts for visual novels or how word count converts to a play time, they get this weird response that is very often "Your VN needs to be as many words as it takes to tell the story!"

Which is, at best disingenuous and at worst condescending and deliberately unhelpful.

Word count obviously matters when it comes to a narrartive-based piece of media. For instance, if you want to publish a novel you will have a target word count when you start, based on what type of novel you're writing. You'll never win an agent over by saying "But I need 200,000 words to tell my urban fantasy story!" or "But I only needed 50,000 words to tell my YA supernatural romance!" 

It doesn't work that way.

So having a target word count from the outset of starting a visual novel shouldn't be that odd. 

It is a relevant part of planning and game design. It isn't the only relevant thing, of course. And technique obviously plays a role when it comes to how the words are being used. But most of the time when new devs ask about word count, they're asking from a game design and planning stand point. So when they're given trite writing advice instead of an answer, it's not helpful.

I spend a lot of time contemplating word count and what my targets mean for the player experience. 

I think about how the words are being distributed through the game as well - so, for instance, if a route is 100,000 words...how much of that is linear (in other words, how much will the player see on a typical attempt to complete the route) versus how much is locked behind branching scenarios and choices.

This is particularly relevant to replayability. 

The way I write my games, we have two types of branching - visible branches where the player knows a branch has happened (for instance, during a choice sequence) and hidden branches where the branch happens without a player's knowledge and they will only have the opportunity to notice if they play through a second time. There are games don't really do the hidden branch thing a whole unless it's related to an ending.

I think indies are really prone to them in general - probably because many indies use the Ren'py engine, which has a clear explanation of how this works coding-wise. And that makes us feel empowered to get creative with branching.

ANYWAY.

In Changeling, the story was about 30% branched on average. If you played every route straight through to the good ending, you saw about 70-75% of the content depending on the route. Some of this branching was obvious because it was directly adjacent to a choice the player just made. Other branching was not as noticeable unless you played through a second time and made different choices that would affect the hidden branching.

For instance, in Ewan's route there's a scene where Ewan and Nora are in the club room together for awhile. And it plays out pretty differently based on whether you have a high romance score with Ewan or not. There's no single choice that will change this - it's based on the cumulative score of all your choices made previously  and the player isn't aware of when the branch actually starts. 

Now, Gilded Shadows (so far) is averaging about 25% branching. GS has way fewer choices than Changeling. I am still working on finishing up Ari's route, which will be the first character route to be complete so I don't have a total number for the amount of choices per route. Changeling had 30-40 per route and we are trying to have fewer choices and options in GS for a couple of reasons I won't get into in this particular post.

But even with the smaller number of choices, the branching is nearly the same amount. This is largely due to the personality aspect of GS, which leads to a lot of small to medium branches where Morgan might respond to something a little differently depending on what trait(s) you've chosen.

Here is an example of that:

I've color coded this so you can better see where the branching happens (just note that the flags could change up in later drafts. This is just the rough version), The grey parts are bits all players will see no matter what. It's the "main" story path. 

If you have the "bold" trait, you'll see the first red bit and the first purple bit only. 

If you are "cautious" and "rational" you'll see the second red and the second purple only.

If you are "bold" and "rational", you'll see the first red and the second purple.

If you are "cautious" and "emotional", you'll see the second red and the first purple.

So, as you can see, this section will play out a bit differently based on different combinations.

The whole section is about 400 words, but the player is only going to see around 200 of that on a single time through the route.

One of my big concerns with the personality aspect of the game is making sure that it's a worthwhile mechanic yet manageable for me as a writer. It can be really tricky to work in these traits because a LOT of our day to day reactions might be quite different if you're stubborn vs cautious, or emotional vs rational.

At the same time, I can't branch every single scenario and

 1. Still have a cohesive story 

2. Not have my brain implode.

That's one reason I tried to phrase the customisation page as a "tendency" rather than a set-in-stone, immutable, consistent, ALWAYS THIS WAY, kind of trait. It allows me a little flexibility to have Morgan specifically stubborn (or cautious) when *I* need her to be one or the other. And then have it based on the player's choice when I think it's a suitable moment for a branch.

All in all, I'm trying to maintain a 25-30% branching average across the routes. There are some chapters that are more branched or less branched, but I'm keeping an on the percentages by tracking my linear vs total word counts. (I have a spreadsheet!)

A lot of devs might not bother to track numbers quite this stringently and that's totally fine. For me, I find it a relevant aspect of game design if not the writing process. One of the harsher reviews we got in Changeling beat us up a little over the routes having a bit of difference in word count. One was around 115k due to the copious branching in that route as compared to other routes.

And it seems this made people feel we were a little careless with our planning or maybe some routes got more attention and care applied to them. 

So this time around, I do want to try to keep things more consistent across routes. Tracking numbers helps.

Anyway, I feel like I've rambled a lot and maybe even strayed from my original purpose a bit. But I did want to give a little insight into the planning that sometimes goes into writing a branched story. This is, of course, just my method, my process, and my general thoughts. Every writer is going to have their own strategy, and if that strategy results in a polished and amazing final product then that's what matters. 

Hope you all enjoyed this little discussion and that is all for now! 

~Esh