Difficulty scaling

[AsHLeX]

Anyone has any tips for making sure that a stat-raising sim isn't too easy, but also isn't too hard - in terms of activating story and character events?

I find that I have a habit of making my visual novels to be either insanely hard (requiring a near-perfect run) or stupidly easy that it becomes kind of pointless. I've added a difficulty setting in my latest sim to try and counter that but now I'm unsure if hard is too hard...

What do the rest of you do for difficulty scaling?

[parttimestorier]

I feel like Long Live The Queen, although it is pretty complex and difficult, might be a good example to look to. One thing I really appreciated about that game was that there were often several different ways to get past skill checks. It's been a long time since I played it so I can't remember all the specifics, but there would be events where someone tried to shoot you with an arrow or something, and you could dodge it if you had high enough agility, but you might still survive with low agility if you had enough first aid knowledge. There were even some checks where a combination of a few skills might help you, even if your individual stats in them were fairly low. Personally, I found that a whole lot more interesting than a lot of other stat-raising games that can end up feeling like you just need to click the button to make the one number go high enough in time.

[AsHLeX]

I've played long live the queen! It was great and yeah, now that you mention it, I do remember something like that happening. Thanks for the tips!

[Zelan]

I also think that it's important to have indicators along the way of whether you're doing well or not. Long Live The Queen might be an exception to this since part of its draw is the huge variety of wacky bad ends you can get. But, in general, I think it's good to give the player an idea of what they need to get a particular ending. I picked up Dandelion years ago and was really excited to play it, but was put off by suddenly receiving an early bad end without warning because the game gave me NO indication of what stats I needed to be raising. It really put me off the game, and I haven't gone back to it in nearly 5 years (lmfao).

RockRobin, on the other hand, I also haven't touched in a hot minute, but I still remember it really well and really fondly because it was so enjoyable. This game gave the player specific goals (i.e. makes 1000 sales by April 14th), so that you knew what you had to do to progress in the story. You don't necessarily have to give the exact stat like RockRobin did (it makes more sense to give exact numbers in dollars or sales rather than like. "beauty" or "talent" lmao), but some sort of indication can be useful (i.e. if the player needs X amount of talent for an event, and they haven't reached that number yet, have an event the week before that goes something like "oh man, the event is in a week..... I should really practice more, I don't feel ready for it yet).

I also think it can be useful to make the "failed" endings interesting. My problem with Dandelion was that I was just happily going along playing the game and then out of nowhere it was just like "oh nooooo the game is over" with not even any input from a love interest and I was like ?????? On the other hand, Natalie by kenibatz (which is currently not available for download, but I played back when it released) lets the player choose how to spend their time on different tasks. There are no early bad ends, but if you're neglecting a part of your life (i.e. hanging out with your best friend or doing freelance work) you'll get messages telling you so. At the end, based on what you put your time into, your friends will be very pleased or very angry with you. It's possible to do very well with one person and horribly fail with the other person. This was a game that I spent a lot of time exploring the endings for, and I actually recently tracked down the developer to get the download link so that I could replay it LMFAO, that's how much I enjoyed it despite how short it was.

Hope this text wall is helpful to you ^^; it got away from me a bit, sorry about that!

[Elfie]

I haven't yet made or worked on any statsim game, but from player's perspective can give you some insights that may be useful:

As already stated above, LLTQ did very well in that regard with it bad+alternative endings, but the best implementation of this I think I've seen in Gaokao100days. Your ending combination (the college you get into and the girl you get) depends on your performance but you WILL get some ending regardless of how good of bad you are. This not only gives player freedom, but also encourages replayability.
IIRC there are also some special options - like for doing certain story choice-based events chains in a certain way, or having some secondary stats such as PE high, or walking in the park every day, etc. So essentially if you delve deep into some side things your stats will matter less (but still will matter). Bad ends, too.

Overall, my points is - a good statsim should not lock the player into hardcore "mathloops" where they will be forced to "minmax or fail". Instead, offer them a good story with a variety of endings.

I may be wrong, but I think most people play statsim for freedom of developing their character themselves and to see how story/ending plays out based on their choices in the game. This way, game is never too easy or too hard - slack off? well, get "so-so" ending. Want more? Well you need to plan better next time!
But, still - never lock stats too hard - because if to succeed in your game, player would need to minmax, it would basically kill all the freedom and posible alternative opotions. When it's like: "go to school event or skip and meet the mystery character in park" it is viable, but when its like: "meet the mystery char in the park or spend day leveling generic stat to not fall behind"(=if in your game options to level up stats are scarce enough for player to be seriously considering whether to see interesting event or level a generic stat ) - it is bad design, IMO.

Because, as a player, if I wanted to just do math, I'd just do my homework or something. Statsims are RPG games first and foremost. And freedom is the key in that.

And adding multiple endings in a statsim is much easier than in regular branching VN. Because, instead of dependency chains and expansive events sequences, you can just add fixed endings based on final stats + whether or not player have seen certain events/made certain choices in them. And adjust them easily, too.

P.S. writing that last one made me actually think about my own current project, maybe I should scrap my hell of a labyrinthine branching and make a statsim instead...

[Zelan]

Overall, my points is - a good statsim should not lock the player into hardcore "mathloops" where they will be forced to "minmax or fail". Instead, offer them a good story with a variety of endings.

This is an excellent tip that makes the point I tried to make in 3 paragraphs in one clear sentence, haha. Definitely agree with this (and with the rest of Elfie's post).