Imagine you're at a cocktail party, and you step outside to get some air. There's a gentleman there who acknowledges you, then resumes his efforts to light his cigarette with a lighter that just isn't working. You stand there watching for a bit, waiting for him to give up. Just then, the lighter in his hand instantly changes to a book of matches. He takes one out, strikes it, and triumphantly lights his cigarette. He then transforms into a pterodactyl, picks up your date with his claws, and flies off into the moonlit sky.
With the exception of my fantastic ending, what I've just described is a Tommy Wonder effect—something he created for moments like these that occasionally present themselves. These moments are so incredibly powerful in part because magic makes a surprise appearance! There is little to no expectation of magic, so the observer perceives the entire event differently than someone with magical expectations.
But, eliminating magical expectations is quite difficult to do for most magicians. For many, it isn't even practical. After all, most magicians are advertised as magicians long before they ever take the stage. Even street magicians, restaurant performers, and magicians who perform informally will often tip their hand when they ask whether a group would like to "see some magic."
This presents a challenge that we've become so familiar with, we probably don't even think about it anymore. That is, a person expecting magic tricks will automatically change what they are willing to believe about what they see and how they are going to perceive it. For example, they might begin to watch your hands more closely, or listen less to what you're saying (thinking that your words are attempts to distract them.) Sure, they may harbor some desire to be entertainingly deceived, but at the same time, they're not going to make it too easy.
Now, it could be argued that some expectation is helpful. After all, they have to pay attention to what you're doing to know something magical happened in the first place. This is true, but the effect is usually not contingent on their knowing that magic is about to transpire, but rather on the performer being able to command enough attention without having to reveal that a surprise is on its way. It could also be argued that some effects require a specific expectation of the magical climax. This is also true, but for effects of this sort, the expectation can be introduced within the routine and is usually not required prior to beginning.
Having an audience without the expectation of magic can, under certain circumstances, increase the effectiveness of the surprise magical moment.
It's a little like expecting a loud noise, but not knowing exactly when it's going to happen. In that state of alertness, you can still be surprised when the expected moment arrives, but you won't be nearly as surprised as you would be if you didn't know it was coming. Likewise, if someone is prepared to see magic, they can still be surprised by the magic, to be sure, but there is an entirely different kind of surprise that is possible when they discover they are already in the magic at, or just before, the climax of the routine.
The spectator in this new situation cannot change their frame of reference like the person who was prepared for the magic long beforehand—there just isn't enough time to do so and no clear point or origin for this mental process. They can either withdraw completely or stay immersed until the tension is released at the end of it all. Either way, they are only left with the option to reconstruct events in their head afterwards. By then, it might be too late, and surprise will have won.
Let's return to our imaginary cocktail party. Imagine it was arranged for the magician in our scenario to be at the event, and he knew that you, the observer in the story, were a central figure among the guests there. Now, what if this magician had not just taken advantage of the moment outside, but had actually created the moment where the two of you would "randomly" bump into each other?
This is a completely different approach from the usual way of doing things and it represents a different class of secrets—the first type being the sort of secret we are all familiar with involving the workings of an effect. This second type, however, is the secret manipulation of the moment itself—causing an intersection to occur between the path of the social situation and the magical path you have in mind. In other words, you are manipulating the environment itself, and everyone within it.
This is an advanced skill to possess, and people who are capable of doing this sort of thing, even if they do relatively standard magic, could be said to have an equal skill set to their colleagues who perform more sophisticated magic, but do so to an audience that is expecting it. Think of this difference like an equation with two sides, where most work on only one side of the equation, and may not realize that it is possible to work on the other side, too. Continuing with this analogy, we can begin to practice this new skill with very simple equations, and over time progress to something much more intricate.
I realize this idea about expectations will not appeal to many performers because of how they structure their presentations and performances, but this is one area where you can have your cake and eat it too. Most of the time, you'll work within the challenges and confines of expectations and it will be business as usual, but sometimes you'll find yourself with an opportunity to manipulate a situation without magical expectations to effect an even greater surprise, and that will be a moment of its own.
Finally, these moments represent a different class of effects. Because it will be difficult to achieve them with certain props that telegraph the idea of magic, e.g., sponge balls, silk handkerchiefs, or special looking cards, the objects one can use effectively might be limited. It also seems that predominantly visual demonstrations with just one or two defining magical moments will work better than complex processes, but that's not to say that someone won't figure out how to make one of those equations work. Every situation is different, and sometimes it just won't work.
But when it does, you have a miracle.