I have now come to a certain point in my design where I have to consider how to balance the player's exitement and frustration. The player in my game has the ability to perform "chain destruction" attacks by making an enemy explode, and using that explosion to cause more havoc. Small description: When a red enemy explodes because of the player's weapon, all nearby green enemies are thrown into the air, falling down after a few seconds on whatever is below them. The player knows where they will fall (small interface widget), and can use his magnetic shield to pull the enemies on the ground on that spot. Falling on a red enemy causes other green enemies to be send into the air, keeping the chain running. The bonuses are high, and as the chain reactions increase in duration, the prime materials (enemies) are used up, making chaining harder, and thus increasing the player's excitement when he manages to chain one more explosion. The problem, however, is that once all the explosions have died down, there's no way to continue the chain destruction: the player would have to start all over again, which is a clear case of the "you die, you lose" syndrome found in some games where failure is associated to loss of power. Now, for short chains, the frustration of losing is small because you can start over again easily, but because of this losses don't mean enough that the player should get excited about his wins. And for long chains, the exact opposite is true, with the price of failure being so high the player is excited when he wins and very frustrated when he loses. How can I find a balance in there?

When the player fails to continue the chain reaction, he doesn't actually lose anything that he might feel he doesn't deserve to lose, does he? It's just that his extra advantage that he had going for a while is lost. But an extra advantage is just that, extra. And the frustration of losing this advantage, I would think, would be necessary in order to make the accomplishment of getting the advantage feel worthwhile.

Here's how I see it: If you make the chaining too easy, then they game will be boring. If you make it too hard, then people will go through the game without even trying too hard to get good at chaining, and thus it becomes a useless feature. The balance I see is between making it too useless and too useful.

It sounds to me like this game has elements that are very similar to the popular web game called collapse, or gem drop, or any that are along those lines. You look to them for examples of how to overcome this. The way I see it, there are two options

Continually give the player more materials until a certain amount have been used.
This can be seen in the way collapse continually feeds from the bottom and gem drop feeds from the top. maybe yours should feed from the side. Have the guys walk on.

Have carefully scripted or set up puzzles so it it possible to use all of the materials perfectly
If your red and green guys are moving, timing might be an important part. let them do levels over if they fail.

I've recently rediscovered Tetris. Awesome game and, now that I'm older, I'm better able to get to the higher levels. How is Tetris relevant? Well, the only real way to get more than three lines (i.e. get a tetris) is to engineer it by filling in and not getting lines and waiting for that stick. Your talking about combos reminded me a lot about this. It can be very frustrating when you're at the top and you're wondering whether you should try to bring it down a few lines (which could block the shaft for a bit, preventing a tetris for a few pieces) or hold out and wait for that stick (but then you just keep getting higher and higher and the music does *NOT* help (which is part of why I play on mute and listen to the radio)). Frustration and excitement. (by the way... did the tetris for the NES mess with your mind withholding useful pieces? I think it was random, and I cringe every time someone mentions making a clone with an AI that screws with you.)

Anyway, the thing is, this kind of frustration/excitement comes completely from the player. They could just get single lines, keeping things low, but also keeping their score low (and it's just fun to see those flashing lights :) ). Getting large combos sounds very much the same. The player gets to control their own level of frustration with how they play. Just follow Agony's advice and don't make them too difficult to get. Also, I wouldn't worry too much about the "left overs", players are used to having to do a little clean up at the end of a level.

Well, from your answers I guess it's pretty obvious winning excitement and losing frustration go hand in hand. What I think would be interesting is making the loss so while it doesn't remove the excitement of chaining explosion after explosion, it still keeps an excitement of its own, rather than a *poof* you're not good at this are you?.

Perhaps considering chaining as a way of "charging up" something. In that case, an early loss would still have an interesting effect, while still keeping the excitement for higher numbers (and even a sense of discovery in the beginning if you add a new effect every 5 explosions).