it is still more effective to kill your opponent before he can achieve another win condition
Very true. However, alternative win-conditions are desirable when no side can dominate the map.
How about an option to have the game end immediately whenever an opponent is killed? For win conditions with cumulative stats, the defeated player might still have more points than the conquering player.
I remember suggesting that collectors or tree gnome had cumulative score, so that a defeated player would keep the score intact and may yet possibly win. My proposal was rejected.
Here I mostly express things that have already been said or implied.
The problem of many strategy games like Widelands is the single-point-of-success. The winner is the player who manages to reach there a little faster than the opponent. Adding prerequisites to that point, it only shifts the rush to whatever is the next prerequisite. [snip]
The more complicated the combination of prerequisites, the less visible this phenomenon and the more chances that an inexperienced player (or some weak AI) may lose his/her initial advantage. But when it comes to experienced players, that phenomenon becomes very obvious and makes the game less interesting. Cause whoever gets the early advantage, will most probably win the game, no matter how many hours later.
Therefore, adding elements to the game to simply delay the inevitable is not a real solution. The solution has to make the initial advantage almost irrelevant to later stages of the game (meaning that if the player fails to translate the early advantage into an early victory, it won't be crucial anymore for a victory after many hours).
I completely disagree with this whole philosophy. If an initial advantage is irrelevant, then what's the point of achieving it? It completely undermines the whole concept of "strategy".
There is especially a false dicotomy here that I point out:
Cause whoever gets the early advantage, will most probably win the game, no matter how many hours later.
make the initial advantage almost irrelevant to later stages of the game (meaning that if the player fails to translate the early advantage into an early victory, it won't be crucial anymore for a victory after many hours)
those two scenarios actually describe the same thing. Even with single point victory, if you fail to translate an early advantage into victory, it won't matter later. If you gain the first supersoldier, but you fail to capitalize on it, then it won't matter in the slightest a couple hours later, when both sides will have dozens.
No, the way single point victory works is that you translate an early game advantage into a greater advantage. Take my latest game with worldsavior at the 2017 tournament: he got a supersoldier 30 minutes before me by some crazy micromanagement, while I built some more the economy: at this point, he used that supersoldier to kill all my weaker soldiers before I could get supersoldiers of my own. At that point I had more production, but he had the greatest army due to inflicting me more losses; so he kept pushing and burned some of my building, took some of my land. Thus my economy was damaged, and he surpassed me economycally, and that's where I surrendered. But his early advantage only mattered cause of follow-through: if he didn't kill my army with early advantage, I would have outproduced him. If he didn't kill my land after killing my army and before I could make a new one, again I would have outproduced him. And the strategy only worked because it was a small map with early contact, and so it allowed to capitalize advantage.
Early advantage only matters if one can capitalize on it by turning it into a greater advantage. If one cannot, then it is not real advantage. Consider a large map where players can't ffight before several hours: getting the first supersoldier is completely meaningless in that situation.
Then you can have an asymmetrical situation with different armies; one has early game advantage, the other late game advantage. but it doesn't change the victory condition. The side with early game advantage must capitalize on it enough to overcome the late game disadvantage. If it does, it will win, otherwise it will succumb.
The only way to avoid this is to increase randomization, which however frustrates those players that are more strategy-oriented.
In the end you only can have two kind of games: strategy games, where you can capitalize on any kind of small advantage and eventually win through it, or games of random where any advantage is easily nullified by chance; most games are a micture of those, though. The closest thing you can have to something that is neither is a sport like football, where having the ball close to the opponent goal is an advantage that is very easy to lose. But football is not even close to a strategy game, and can still be modeled as a strategy game where having correctly advanced close to the goal gives you a higher percentage of scoring, after which if you fail the random you will lose the temporary advantage. Or you can model it as a succession of many mini-games, where every time the ball is stopped the old game ends and you start what is by all accounts a new game. And the victory goes to whoever won the most mini-games.
So, for somebody who likes to see his skill and investment rewarded, there is no alternative to single point of success. You make it sound like it's bad, but it's the only kind of game that rewards strategy and skill. The way to keep it interesting is to make it very difficult to capitalize on an early advantage, so that it is perfectly reasonable to lose it. And widelands does it well, I think. You are putting so much stress on getting the first supersoldier, the strategy by which I started to dominate, the strategy worldsavior improved and defeated me with. You see two people defeating everyone else with that strategy, and you think it's all about micromanagement of the fiirst supersoldier. But the economy tournament showed that we are also much stronger than everyone else at making an economy, with worldsavior having a small advantage over me even there. It's no coincidence. Getting a supersoldier first is but the result of setting up a full working economy first. Skill at making a supersoldier requires skill at economy.
So, delaying the supersoldier with my proposal would not change the single point of victory principle, because that's inherent to any game with strategy, and it's not undesirable. It would however have the following positive effects
more delay for the victory means more chances for the advantaged player to make a mistake, which is the closer you can have to randomness while still rewarding skill.
more delay means an easier time balancing the tribes. getting the first supersoldier 10 minutes earlier matters much less if the game is already 3 hours long.
more delay means no chances to skip segments of the economy to arrive faster.
more delay means a time to use a feature of the game (half-promoted soldiers) that would otherwise be ignored.