The topic of suspense came up in another thread, and to avoid derailing the topic I will attempt to open it here.

A little reading on suspense as an audience experience suggested the following:
Audience needs sympathy with the characters
The consequences must be significant
There was a suggestion that it enhances the experiences if the audience knows more than the characters do about the situation

There was also discussion of probability of success related to the feelings of suspense:
Good outcome is very low - you are hoping against the odds
Good outcome is very high - yet we are watching, so perhaps there is somethig in the works to foul it up
Good outcome is 50/50 - literally we can't figure out which way it will go, most uncertain

I believe that all of that is decent information to think about in game player from a purely number crunching game system to a game depending purely on discussion without elements like dice or stats. It helps you set up something that will feel right.

My question is can a more mechanical approach enhace the experience at all? If so, how? What part of the players minds is it aiming at?

The first approach I ever saw to this was in Masterbook/TORG. Players always have a hand of cards from a game deck where each cards has values that cover action order, random modifiers to given action type to encourage variable tactics, and a Sequence mechanics with letters A-E. If there was a sort of race against time, the players had to draw and play in order the A-B-C and so on sequence in a certain amount of time.

For a game not using the deck, I could see you stating that you have N number of action counts (be those rounds with a set time, or more abstracted actionr resolutions) to succeed at whatever.

Another approach is to use a deck of cards, or dice, and say that each round of resolution a card will be played or a die will be rolled, and when it surpasses a given total, the end resolution has happened.

So for instance, if my players are trying to get to the escape flitters on a dirigible that is going up in flames, every round they navigate the smoke filled corridors and fight the nefarious Men in Brown Bowlers a card is laid down. Probably face up. And using the typical Blackjack values,  the players see when the card is a low or high value, but they don't know what the total I am looking for is.

Same idea, but you roll a die.

I personally am attracted to the idea that you lay out the cards, like ten, and flip them as action proceeds. Why? because I am phobic of the feeling that I personally will have to decide that they have failed miserably. Or that i let them get away with it. Or that they will feel that I let them get away with it.

Mind I am fine with a system that lets the players use up some reource they have to remove a card or gain extra time, or even to convert a failed effort into a marginal success, just as long as there is a cost that could have been avoided if they had had better luck or had better skills or better decisions.

Are there other ways? Anyone have any experience with this sort of situation, and does it work to make the players feel more involved? Is it based on bad assumptions about how suspense works, or can work?

Since I am starting up my Victorian Adventure game

What came to my mind upon reading the title of the thread are extended resolution tests.  Say two characters are in a conflict of some sort.  Ten resolution tests will be made, but only the winner of the final test wins the conflict.  This way there is a certain amount of suspense right up to the last moment.  Say each test has a 50-50 chance of being won by a single character, but the winner gets a +20 on their next test, making it a 70-50, or 58.3-42.6.  A win increases the chance that the next test will be won, increasing the likelihood that that character will have the bonus for the last test, and therefore the greater chance of winning.  But one loss and the odds have completely reversed.  Who will win in the end?  Even with the bonus, victory is not certain.  Thus there is a certain level of suspense right up until the numbers for the final test are known.  If the stakes are very high then the suspense increases dramatically.  With this sort of thing it is important to avoid the "death spiral" effect.  If you can tell who is going to win after the first three tests, then there is no point in doing the rest and the suspense is killed, hence why I said that each success only applied a bonus to the next test, rather than all subsequent tests.

I'm sure there are a number of ways to create suspense via mechanical means.  This is just one way that came to mind.  A more general way, and probably similar to some of the things you mentioned, is to let the situation depend on some random generation method (like dice or cards).  Say the players draw a Noise card, that indicates that they heard a noise somewhere nearby.  They have to decide whether they want to Encounter the cause of the noise or play an Evade card to avoid it.  If they Evade then they can investigate the noise (giving them a bonus if it is hostile) or ignore it and continue.  No one has an Evade card so they roll the dice and consult a table: the noise was a savage band of orcs!  You can create a similar level of suspense by inventing such a situation, but I'd imagine the effects are somewhat different between knowing that the GM is controlling every thing and knowing that nothing has been decided and the GM (if there is one) is waiting to see what happens just as much as you are.  In this case I'd say part of the GM's responsibilities would be to come up with new charts so that the group doesn't encounter the same thing time after time until it gets old.  In fact, this is giving me an idea for a game system.  Hmm.  The cards would have to be somewhat generic or else certain combinations just wouldn't make sense.  For example, a Cave In card would only make sense when the party is inside a cave, or at least some kind of building or someplace where there is a roof above them to fall down.

I think suspence is often invoked in movies or TV by something morally reprehensible potentially happening (or atleast what most people would generally find horrible). A recent example I can think of was in the comedy 'Lead Baloon', where the old ladies cat is in the garage (where the guy was supposed to have searched) under the car, and he gets in. They kept cutting back and forth from him fussing with car stuff/getting ready to the the cat staying very, very still right next to a car wheel. It was rather effective, let me tell you!

BUT perhaps the shocking thing with roleplay is - well, with TV, I don't know which way the writers going to go. Not fully. But in roleplay, were the writers - we know which way were gunna go. No suspence, unless...

We use dice to determine it (or roll to determine who narrates it alone, which runs off the same suspense as not knowing the writer from above)

But that means the cat can really die. Just crushed to death - not in some grand drama, but as a stumbling accident. A nice little cat just dies - no saving moment. Nothing. The really horrible can happen in roleplay.

So you can't just have suspence in roleplay just for a bit of a thrill - there will ALWAYS be a genuine moral threat involved. Play cannot be safe and harmoginised. Do you want a game which does not play it safe?

That is if moral danger is needed for suspence. Might be wrong but I can't really think of any other ways.

I will post some personal experiences, a research review and my design related thoughts here when I find the time to. I think this is a fruitful topic.

Thanks for the replies, and I look forward to hearing more.

One thought I had is that maybe there is more application to this sort of mechanically enhaced suspsnes than the obvious countdown types of scenes. I would lvoeto have players setting their stakes, and investing more and more effort into something, but with some mechanism hidding the final results, or even the full value of how well they are doing at the moment. IF that enhanced player involvement and suspense. At some point it has to move off into how the setting reacts to their actions, but I wonder how far you could take it.

A simple example would be meeting with many groups in a political situation, and accumulating points with the groups towards supporting your goal, but those points eventual value is uncertain. Again you might say each point is d10 votes, or what have you. You just know how many dice you have gained not the final results of said dice. And you could place secondary actions into setting minimums on some dice, similar to the special dice in ORE.

I am thinking that while I truly am attracted to very atomic procedural type play, in the end I want something that is building towards a large scale of resolution than some small task provides.

Chance,
 I have been thinking about it and my own play. And it SEEMS to me like, if the stakes of the roll are important to the player, there will naturally be suspense, but if the stakes are not there, there is no suspense.
  Like, if your character is walking down a path and the GM calls for  a perception/awareness roll. There is no suspense. Because between the 2-5 people in your group, someone will make the roll. It doesn't matter if you, personally, make the roll. As long as someone in the group does.
  But, I noticed when I did negotiated Stakes in TSOY, there was a lot of suspense on each roll, because I dialed the stakes of losing into what was important to the character and player.
  For instance, there was a character that had Key of the Guardian and the teenager they were watching over was being a jerk. And I had a character try and kidnap her. One of the group members says, "I rescue her!" So, I made the stakes of failure be, "If you fail, you rescue her, but your bodyguard dies." One of the players piped in, "Shouldn't the stakes be, that she gets kidnapped?" I replied, "No, because you guys want her out of the picture."
  Then everyone leaned in to see the roll because they cared about the stakes.

  I think this illustrates my point, but I dunno. I never really push mystery or suspense too hard. And that usually makes for a good game, because I am not trying too hard.

Check out Dread!

http://www.tiltingatwindmills.net/dread/index.html

Also, Callan's observation of how suspense is often created by having a bad outcome or danger visible to the audience, but not to the protagonist--that's interesting.

That would be hard to recreate in an RPG. Has anyone/any game done this successfully? How?

Oops, I forgot about that! I meant it as an example even for when protagonists know a morally horrible situation is at stake. The same issues apply. I didn't mean to go into some complicated audience knows/PC's don't thing, even though that's a fruitful vein to follow as well. The base of either still requires the morally horrible. And as I noted, in movies/TV the author can never intend the kitten to die, and yet there is still suspence. But in roleplay, with dice, there has to be the genuine chance of it to generate suspence. You can't have safe entertainment like with TV/movies.

Paul wrote:
Also, Callan's observation of how suspense is often created by having a bad outcome or danger visible to the audience, but not to the protagonist--that's interesting.

That would be hard to recreate in an RPG. Has anyone/any game done this successfully? How?

To answer the topic,

Can you create suspense via mechanics

I'm getting the sense that in games where a lot of things are resolved by a Fortune Roll, the very act of rolling dice (or playing cards, RPS, etc) adds tension.
The hard part is that it's hard to suprise the player(s) responsible for the plot. 
In more GM-centered games, for example, the GM always knows what NPC is betraying the PCs, because that's a scheduled event.  In less GM centric games, it's even harder to suprise people.

I would suggest some sort of Mechanic where a Player pays some sort of resource, and other players can determine how that resource might be used. 

Now, imagine a Betrayal pool.  If it's on the PC's side, the PCs are betrayed.  If it's on the GM's side, the GM gets betrayed.  And unavoidable actions (like any combat skill check?) flips the side back and forth.

For example, 'Alex' gets 3 points, and puts it into the GM's Betrayal pool.  He's hoping to burn that pool and have an NPC betray the emeny at some time later.  But 'Betty' uses her 'Double Damage' power to get through a current conflict.  That flips the pool over to the Player's side, and the GM can now burn those points instead of Alex.  Alex needs to hope that the GM uses a power to hurt Betty that flips the Betrayal pool, or that he can figure out a way to flip it himself.

Alex feels suspense, because when he paid into the pool, he didn't know if he was making a good decision or not.  Betty has risked the Pool to end the conflict quicker, so she feels suspense, as she is risking the group's potential betrayal.  The GM doesn't know who is going to betray whom, so SHE feels suspense.

A downside is that when someone actual burns those points, someone else ends up being betrayed.  In a GM-centric game like DnD, the plot may not be that flexible. But in a game like DitV, it might!

-Maybe something like that?

Oh, I got more. A lot more, but this I want to get off my chest now.
http://www.springerlink.com/index/9t518024t8427m33.pdf
This is an article about a chomputer program for suspense story making.
Not quite as fascinating as it sounds, but still a nice read.

Another thing I find to be highly interesting: Oral Narrative Traditions.
Before we focused our "literature" on, well, writing, we told stories. This was actually a lot more similar to RPGs than I first thought. The narrator typically worked of a framework that was known to everybody. Like, the room told the narrator to, again, tell the story of how Gilgamesh fought Enkidu. Everybody knows how the story ends, and still, the suspense is deep. Becuase they knew who won, but not how. This was the narrators job to make up.
This is really similar to some FitM conflict resolution rules (say, otherkind).