I vaguely thought of this when I first watched the Serial Killer Roguelike character generation video, but it was this
Expedition video here that really brought it home:
When you're advertising a roguelike, you want people to want to play it for the
first time. It's good to have something going on that makes people expect they would want to keep playing, but the only thing you can do to the vast majority of your potential audience is to get them to play for the first time.
Watching the Expedition video, I felt that if I put some work into that game I could really enjoy it. Finding just the right optima between risk management and growth potential, learning how to navigate reliably, exploring interesting new lands, squeezing out the most gold that I can, what's not to like? It seemed like it could be a great game on the tenth session, and I would have liked to play my tenth game of Expedition. But I
don't want to play my
first game of Expedition. Or the second or, perhaps, even the third. I don't want to play them because based on what I saw in the video, they'd just end up with me choosing some poor balance of resources and dying ignominiously before I get to anywhere interesting.
Compare to SKrl. In SKrl, I could choose a character with high stats and mild or no disorders (assuming that the game's challenge level is not affected by the stats you choose, effectively making your stat choices a reverse difficulty setting), and the game would dump me somewhere where I could go and kill people. The videos made me want to play SKrl for the
first time - even if the game could turn out very difficult, it never seemed overwhelming. And even better, the video let you see enough that you knew you would likely still want to play another game, and another, down to trying to survive with an all-minimum-stats and heavily-messed-up-in-the-head runt.
To see how the SKrl videos went right about this, consider how they could have gone wrong. Suppose the game had had the precise same feature set, except that instead of just choosing your stats, you would have had options like "go to gym for strength training", "play ball games for perception and agility training", etc.. Each with a dollar cost, of course. Oh, and for play balance, a "have traumatic experience" option that would give you a negative trait but refund you a certain amount of dollars. All of that would have seemed... pretty overwhelming and pointless. And it would have annoyed those who haven't yet reached enlightenment on the realism vs. gameplay issue :p. (And if you tried to bring it closer to the "pick your difficulty level by picking your character's level of capability" by letting the player choose how much money they start with, well, that would just have exacerbated the pointlessness)
As a final illustration of the point I'm making... if SKrl is "good" and Expedition is "bad" in the I-want-to-play-this-game-for-the-first-time sense, Prospector is a pretty good example of a game that's "eh, close enough". The ship choice menu has an obvious choice for the first play, based just on the name of the game (you probably would want to prospect with a scoutship!), but also shows based on what kinds of ships are available that there's more to do in the game than just prospect. The amount of money it gives you is enough to make some interesting choices at the first station once you're more familiar with the game, but small enough not to feel like much, like you could mess up your first try by not buying something essential. You don't have to worry about fuel before you've left the station, you don't have to worry about how to repair your ship before it's actually been damaged. In short, Prospector doesn't feel too overwhelming, and a lot of the things it does do say "I have potential!" to the player. It's just that the SKrl videos did the same sort of things even better. Killing people as a horrifyingly ugly and clumsy schizophrenic, for instance, just sounds like a more interesting challenge to eventually get to than going from prospecting planets to trading contraband does.
