Damn near everything doesn't work in every set of conditions. If I put your car on an icy lake with slick tires, it ain't gonna go.
If you put your tires on dry pavement, they will have good traction every time. If you put them on wet pavement or ice, they will have less or no traction, again every time.
If you try to lift off in your airplane in the middle of a thunderstorm, or loaded full of gold bricks, it ain't gonna fly. That doesn't mean it DOESN'T work under every set of conditions.
However, if you load your airplane sufficiently above gross, it will fail to take off every time. Perhaps you are coming to a smallish understanding. Its called reproducibility, and it is the cornerstone of just about everything we understand about the world around us. Of course there are conditions that affect the properties of energy and matter. These conditions are observable and, if understood, predictable.
Besides, the whole "prove it" thing is reminding me of an elementary school playground. There's more than one person who says it has worked for them, so there's probably something to it - And they must have encountered the conditions where it works. Your assertion is that it never works - Why don't YOU "prove it"?
No, you are mistaken. My assertion is that it works at the same rate as random chance, because that has been the result of every controlled experiment that has ever examined the phenomenon.
Now, if you tell me that your friends are really astute, and can see the effect that underground water has on the vegetation or geology above it, I would believe you implicitly. Those looking for the water might not even explicitly know what they're looking for. But if underground water can effect what's above it, I have no doubt human beings could deduce that pattern.
If you tell me you think people can locate water, oil, gold, the lost city of the Incas or anything else by watching a couple sticks, rods or whatever, then I have a nice bridge to sell you in Brooklyn. I have no doubt this method is as effective at finding these things as random chance. And as the saying goes,
even the blind squirrel gets a nut. It's an old saying, lots of people have said it, so there must be some truth in it.
What I meant by "prove it" is really quite simple. There's a million dollars out there for anyone who can prove they can really find water with just the sticks. I doubt Randi will ever have to pay for dowsing or anything else, but because there have been so many yahoos and charlatans calling, you now need some credentials to give it a try. The kind of credentials I have. I will design the experiment so that other scientists will believe its result. I will stand by the result whatever it is; if the experiment is properly designed and executed, you get the truth. And if it turns out these guys can really do this, I will do my UTMOST to make certain they become known in the scientific community, and that they have a fair chance at that million dollars.
Sorry, if I had that much confidence in my ability and an offer like that, I would go anywhere, even Michigan itself, to do the demonstration. But I could use a million dollars. I guess all these dowsers are independently wealthy or something.