Hmm, take a cue from literary theory? Tough twists build tension. So let the tension build and build and build with increasingly brutal twists, then after the climatic moment they are in the falling action. You no longer want tension, you want excitement. Will they get away with it, and how much will it cost them? So instead of increasingly brutal twists you apply narrow escapes and more conditions. If you’re really good you’ll mix these together and get that rollercoaster feel, with little climaxes and falling actions throughout the session, each climax even more tense than the last, each narrow escape more harrowing but relieving than the last… until they reach the true climax, flee the scene, and finally step out into the sun. The walk home and town are your denouement.
One of the things I found out during play (and a feeling the book gave me after re-readings) is that Twist > Conditions.
As jovialbard says you want to build tension (with twists), then the time comes for a tough decision (do I let go the treasure sack to climb on? Should we flee to the unexplored room? Should I try to read this skin-bound book we just found?). If the dice are gracious then the players will triumph. If they don’t, then apply the Condition and the players will endure. You don’t want the characters to die.
I had the opposite management the first sessions I GMd, letting them lead the action while applying conditions. By the end their characters were completely useless, and the risks had become deadly. If I had placed more twists, they might’d being in greater danger, but with more possibilities to confront it.
Stay cool 
If you get a condition from attempting to leap across the chasm, then you have successfully leaped the chasm. You may have twisted your ankle when landing, but you’re on the other side. As jovialbard notes, if you fall in, that’s a twist and you might get a Health test to avoid injury.
DagaZ, that’s gold buddy. I did the same thing. I applied conditions on everything at the beginning of the session and let them push everything their way. I’m going to try it the other way around now and try to set up a Twist, Twist, Condition, sort of rhythm (staggered, of course, but you get the idea). Eager to try this out now.
I’m preparing to try it with my group. I guess the first one get’s to play can post the results! 
Stay cool
I agree! I am also looking forward to trying this in my next session. I feel that I have been imposing too many Conditions and not enough Twists, and perhaps that’s just because Conditions are “easy” (for the GM, not the players). Time to stop being so “easy” and work more towards being “interesting”.
Not trying to circle back too much, but I have another workshop-type question in this same vein. I want to know how hard you would twist in a case like this:
The adventurers are looking for tracks their enemy may have left so they can follow them. You call for a test and they fail.
You think to twist it this way, that they found tracks but they’re an Ogre’s tracks. But, the question is, how far do you go? Do you:
-
Not tell them they are Ogre’s tracks - just they they are “tracks” - and let them decide if they follow them.
-
Take it a bit further and tell them they found the tracks and followed them, bumping into what first seemed like a stump, but turns out to be an Ogre’s ankle.
What do you think? I like twists where the GM takes control for a second and gets you right to a problem, but some folks seem to think it should be broken down in a more granular way. They think the adventurers should have a choice of whether they follow the tracks, especially if their intent was to simply find tracks, not to follow them. But even when their intent seemed to be to follow the tracks, they still seem to think they should have a “saving throw” so to speak.
In my opinion a proper twist causes trouble, so bumping into the Ogre or at least having the Ogre be nearby, noticing their presence and stomping in their direction would be a proper monster twist in this case.
Trying to figure this all out and improve my GM style. Thoughts much appreciated!
If they say “We want to find if they left a trail and follow it” then you are perfectly justified in #2. They found tracks! Not the tracks they were looking for, and since they couldn’t tell the difference, they did what they said they were going to do and followed them. They can’t back out once the dice roll.
If they say “Is there any sign that our enemy came through here?” then I would not do #2. They haven’t said they want to follow the tracks, so making them follow the tracks would be wrong. Reward players for being cautious and careful when they describe what they do. However, you can say that they waste time sifting through the bushes and trees in the area, and in the meantime an ogre stumbles upon them, or a pack of wolves sniffs them out, or something in that vein. At least in that case they have more options than they would when walking right up and bumping into the ogre’s shin, which is good, and that’s what they get for being cautious adventurers.
You could also tell them that they found tracks and not tell them that they are the wrong tracks, then wait to see how they respond. But you’re right, that isn’t an ideal twist because it isn’t an immediate problem. There’s nothing wrong with it, it’s just not ideal.
I agree with jovialbard.
Also, you COULD save your twist for a little later on (like the last example of jovialbard) with the addition of stating that those are the only tracks found. This would introduce a two-folded problem:
1st) if they follow the tracks they will find an ogre instead of the intended target.
2nd) by stating that these are the only tracks they could find you are implying that the actual target never left noticeable traces of their travel.
#1 is a future twist and #2 is an immediate twist. A clever player might inquire about the tracks and you can tell them that these are ogre footprints (or make them test), and/or conclude that this is not the trail they searching for and look for another way to track the target down. A more brash player would follow the tracks and discover the ogre at the end.
Of course during play this is hard to tinker. As I see it problems are about introducing new, harder choices for the players (do they stay and fight this monster to continue through this tunnel or do they flee and look for another way around?).
Stay cool