Regarding "story", my approach is not to start a campaign with saying "there are thousands of people in the world who want to become heroes, even though most of them fail. Let's see what happens to these 4 randomly selected people".
Instead I start with "Of the thousands of people who want to become adventurers almost all fail. Let's see how these 4 most promising of them are doing."

The outcome of their adventures is still entirely in the open. They might still all die on the way to the first dungeon, depite all the promise they had been showing.

A novel or skript writer of course does know the outcome of the story and therefore knows exactly which of the millions of people in the world are the most interesting to follow around. Which any decent GM would not.
But I like to increase the odds that the campaign we're playing is with one of those fee groups who make the jump to becoming heroes. I want to minimize the chance that everyone is playing five or six Basic Level characters before the group finally makes it to Expert Level range. Playing at Basic levels is fun, and that's where the vast majority of sessions I ran took place. (With several different groups.) But once a group of players has played at those levels for 10-20 sessions, I want to progress to higher levels and do different types of adventures instead of going through several PCs per player before we get there.
And I think with B/X as written, the odds for that are much lower than I would personally prefer. Always being vulnerable is good (and even more than the workload for GMs, that's my main problem with 3rd edition), but being that extremely vulnerable at 1st to 3rd level is too much for my taste. Maximum hit points at first level, 3 or 4 spells per day for 1st level wizards (even if that includes scrolls), ranged weapons for wizards, thief skills that start at 50% chance, and no instant death traps when trying to identify magic items (with a 1 in 8 chance!) is really all I am asking for in this regard.

But I am interested in what fundamentally different concept of sandbox campaigns hamlet might have.