I have a game I idea that I just now thought of so its very rough. I have setting and scenario but I'm trying to decide why everything happens and what the goal is. Is there a case where its acceptable to have an undefined win scenario? I figured that it might make it so that there wasn't a clear end point to the game without one making it possible unsatisfying. Thoughts?
Win Scenario
If you want someone to "win" your game, then you'll want a predefined end condition. Otherwise, what you have is a sandbox: a place where players simply do stuff.
The big difference is in giving players a reason to do what they're doing. Is it just a sandbox? Then cool, there's no pressure. But if there's an end-game condition - if someone is supposed to win the game - it will likely influence players to choose one action over another. This is typically because if someone is playing a game with a goal in mind, they'll likely choose an action that brings them closer to the goal, as opposed to simply doing whatever they feel like doing at the moment, or whatever currently piques their curiosity.
For example, early versions of Minecraft were considered a sandbox. The players simply interacted with the world. Sure, they dug through caves, crafted items, and fought monsters, but there was no specific "end" to the game. Players likely did those things because they thought it was interesting to do (or maybe it was because if they didn't defend themselves from monster attacks, then they'd have to start over from scratch).
Nowadays, there's apparently the Ender Dragon (or whatever it's called). Once players defeat it in battle, then it's considered that they've won the game. There are lots of ways to go about it, plenty of action to be had and choices to be made before then (and maybe after?), but at least in current versions there's an end-goal in sight.
I'm not personally a Minecraft player these days, but that's what came to mind when I considered your question.
Well first, of course context is key. Since I don't know the setting etc. it might play into my response. Either way, generally, I think that some kind of winner is good. People like to strive competitively for that win! Also, what you could do is have multiple ways that you could win, thus there might be strategy to different ways you could play the game. You could have multiple winners, maybe, but I think some kind of defined end is good, it gives closure. But, again, it depends on the game and I would be open to hearing what this is all about :)
-Austin
Whether there is a win condition or not, there should be a way for the player to experience accomplishment or progress.
Many games have fixed lengths or single storylines with an end or a set of conditions where the game ends. Then there would be a chance for scoring. In a competitive game players compare their scores to each other; in a cooperative or solo game they might compare the score to a chart of ranges to get a sense of how well they did.
Even in a sandbox game, players will attach significance to milestones and may even refer to them as "winning." You might as well make it explicit, even if you want more than one way to win.
It's relatively easy to spitball a few ways of "winning" a competitive game, the difficulty is in balancing them so no one becomes clearly dominant.
I'm a big fan of personalized goals that encourage different play styles, but not secret goals that lead to sudden ends.
Even if you go for the bog-standard Victory Point track, player/avatar differences can affect how many Points are earned by a given accomplishment. The Warmonger gets an extra point for each battle won and no points at all for mediating disputes (even when it's in her interest), while the Utopian gets an extra point for each major step on the research track and no points for suppressing rebellions (just keep the citizens happy), etc.
Conversely some games (like Jenga) simply have a "lose condition" (i remember they talked about this on ludology during the episode defining "what a game is").
As best i can see you could have lose conditions, end conditions, or win conditions; but in order to be a game you need one of them so SOMEONE (maybe a single player with hidden info or a dungeon master) knows when the game ends.
Those are good points. I hadn't thought about it being a sandbox. I'm leaning in the direction of having multiple possible scenarios and multiple goals depending on the players character, at which point it becomes a problem of outlining them all. Another thought that runs along the lines of a sandbox is having a choice of whether or not you want to go for a win scenario. Would that make it pointless?