Most of my time on these forums is spent answering questions like these, so I thought it would be a good idea to put them all together in one post and never have to worry about them again. So, let's proceed.
What's a good trick to blow the audience away.
The short answer is: there isn't. The misconception here is that tricks blow people away. They don't. Magicians do –
with the tricks the perform and how they perform. The best way to show this is with a couple of videos of the same trick, in this case the classic Balducci levitation. One gets a good reaction, the other doesn't.
Video 1
Video 2
So what's the difference if it isn't the trick? Well, it's the presentation. To begin with, the performer in the first video is not the kind of guy who should be able to levitate. In order to levitate, the magician should appear physically fit, and as if he has studied and practised this sort of thing for several years. It's no use just saying that you have, you most certainly have to appear so. In this case, performer #1 most certainly fails, while performer #2 succeeds. For the same reason, I loathe when young magicians try to copy Criss Angel's repertoire. Angel actually looks as if he can do what he's doing; children don't, and they're going to get torn apart because of it.
So, it's about
how you perform the trick that blows the audience away, not the trick itself. I have performed the basic key-card trick to people and "blown them away".
Is [Trick X] Good/Worth Buying?
It is extremely naïve for people to thing that there exists a simple algorithm for separating magic into two piles: good and bad. It's like with anything – why are there so many different models of mp3 player, designs of houses, etc? The answer is becasue they suit different people in different ways.
So, we can alter the phrasing of the question to give: "Does [Trick X] Suit Me?" That's a little better, but raises the question: "Well, that's up for you to decide since you're the one who'll be using it." Despite that I'll do my best to help you answer the question for yourself.
When you're looking for a trick to buy, look first at the effect, independent of all the conditions for performing it. Does the effect suit your character? First and foremost you need an idea of what your character is. You don't have to have it worked out to a tee because it will evolve over time, but you need to have a rough idea.
As an example, being a computer scientist, my character is based on logic. I wouldn't perform a trick without giving a scientific explanation for what I'm doing, no matter how ludicrous. If the rubber bands are passing through each other it's becasue the atoms are aligned just right so that those in one band are going between the gaps of those in the other. Therefore I ignore tricks like levitations that rely more on spiritual phenomena.
Sometimes the effect as written will not suit your character but you may be able to adapt it so that it does. Oz Pearlan presents Killer B as a prediction excercise. I use it as a demonstration of the power of NLP and suggestion. This aspect of adapting a trick is a very good topic on which to ask questions. There have been a number of times I have seen a trick, thought about adapting it, bought it and realise it can't be adapted in the way I was thinking. So, I'd always recommend asking questions like "I was going to buy [Trick X] but I was wondering if it could be done this way instead of that."
Ultimately, you shouldn't be able to expect that others will make your decisions for you. This is about what
you want, not what others think you want. If you don't know what you want, do some searching. Browse magic sites. If nothing grabs your fancy you may want to consider and alternative hobby. If too many things interest you, either try to eliminate some using the process I've mentioned above, or pick one at random and buy it. You can always buy the other(s) at a later date.
That's it for now. When I come up with other questions dealing with misconceptions I'll add them in.