They take advantage of a design flaw in most pin tumbler systems called MACS (Maximum Adjacent Cut Specification) that limits the height difference that two adjacent pins can have.
I understand what you're saying, but I don't think it stands up to mathematical analysis. Without spending all day on the assumptions, I would estimate that these picks are equivalent to trying about 250 keys. Which just isn't enough. When you think about it all those 'carefully' contured shapes can only be effective as a key (ie you're not scrubbing), when they are pretty much all the way in, else they aren't lifting the back pin. If they have been 'scientifically' contured then they must also assume a pin pitch, which will in practise very from lock to lock. If you are going to use these as artificial keys, then you are going to have to apply torque once for each possible simulated pin combination, and then release it. And you would have to do this some hundreds of times. I bet that successful user aren't doing this.
I would bet that the reason they actually work is that they are unwittingly being used (quite gently too) as rakes during the insertion process, and then a little more when moved up and down.
Still, I haven't tried them, so this is only theory, so feel free to shoot me down in flames.
