I think you have to take into consideration how large the setting is relative to the player. For example, the game Harvest Moon: Save The Homeland. I'd describe this game as a freeroam RPsim (no combat) with a small setting (one village which will soon cease to exist unless the player puts work into preventing this. But there are actually about 9 different ways to save the village, which are more or less separate quest lines that run in parallel for the duration of the story. Almost all of the freeroam quest content in the game advances one or more of these village-saving quest lines. It produces a really pleasing feeling that everything in the world is connected, and there isn't one artificial main quest plus many tacked-on side-quests.
Whether that could be done in a setting as big as Skyrim's, well it depends what sort of goal the game has. If the main goal of the game was something like mastery of one career or type of skill, that would be relatively easy to distribute so the player has to go to every location in the game once to collect all the subskills or training opportunities. If the goal is even more basic, like amassing a target large amount of money for some goal presented in the starting area, that's very freeroam-friendly, if not very strong narrative. Or perhaps a game has a central location, perhaps a school or military base, which is defined in the story as the player's home base, and the player is expected to return there each time they want to advance the main plot (presumably after racking up enough freeroam play outside this location to earn/unlock that next advance in the main plot.
One alternate game structure I've looked at personally is a game split into days or chapters, where the player can choose one of the locations within the game to spend each day/chapter at, then travel to a different one next. The main plot proceeds for N days/chapters, regardless of where the player is, but the location affects which options the player has to influence what happens in the main plot for that day. Additionally, all locations have something that can be achieved once before the end of the game, so the player is encouraged to visit each location at least once.
If the quests are story-focused, make them as long as they need to be to tell the story. If they are side quest just there for fun, make them as long as you think the player will still be engaged in them. If side quest aren't engaging then they are more of a hassle than they're worth.
Then the question might be: why are you creating a story-heavy game in the first place?
A game isn't likely to be a good fit for all players: some like lots of story, some like lots of exploration, some like lots of combat, and so on. I really don't recommend attempting to make a game that doesn't lose any players; instead, I'm inclined to recommend making a game that you yourself might enjoy, whether that's story-heavy, sand-boxy, gameplay-heavy, or whatever.
Each story has a sort of general result of the main character in each making a choice to change the world fir their own personal reasons. I don't want to go into too much detail but I have 5 games planned in the series and they all trace back to the three stories at the beginning.
I don't think there's a single, best Main Quest : Free Roam ratio; I don't think I even have a favorite. It just comes down to what you want to do with your game.
The Elder Scrolls and Harvest Moon games tend to have very low ratios as a natural consequence of their emphasis. TES is about exploring an expansive open-world, and so the questlines direct the player to a wide variety of locales, before letting go of the leash and letting them discover things on their own. HM derives a lot of its fun from the player losing themselves in the daily routine of farming and fostering relationships. For this reason, quests can often be completed without even doing anything differently.
Deus Ex has a much higher ratio, because it emphasizes the ability of the player to express themself through play. This requires a more linear path through the game so that the designers can ensure that being a diplomat is always feasible, and so is being a thief, and so is being a killing machine.
If each of your games is to be about the consequences of a single, pivotal action, then you want to make those consequences palpable in as many ways as you can. Many (though perhaps not all) side quests should deal with the far reaching effects, and if you can make it work, the entire game-world should change over time to reflect the unfolding results. You might want the main quests to punctuate the change that the player has brought about, highlight certain things, and encourage the player to think about their action in new ways.
If the story is a core element of the experience, then why worry about players who don't want to experience the story? Isn't that roughly analogous to building an RPG and attempting to please players who don't want to level up?
If the story is a core element of the experience, then why worry about players who don't want to experience the story? Isn't that roughly analogous to building an RPG and attempting to please players who don't want to level up?
The idea is to make the game as enjoyable as possible without letting things get too muttled or sacrificing what i want the world to be
The idea is to make the game as enjoyable as possible
Enjoyable to whom? There are players who like linear games, players who like sandbox games, and players like me who don't care one way or the other as long as the level of openness contributes to the game as a whole. When designing a game, every decision removes possibilities; the final product is the one you leave behind.
