Again, you aren't describing science. What you ARE describing is anomaly. One off events that, yes, can happen, and because of Quantum events, should. But I will refer back to my gasoline metaphor. Even if I conducted the experiment 1 million times and, anomalously, it failed to ignite 1000 times. I would not presume that to do it a third time (or a hundredth) and get to 900,000 consecutive ignitions that I am approaching a "non-ignition" event and then stand in the pool. I would try to better understand the anomalies because there is SOMETHING I missed, the overwhelming odds scream of it. I will even relate a story of something I witnessed that at first impression seemed to be a UFO. I spent time in the Gulf (Desert Storm [manufactured Middle East debacle number one, at least in my lifetime]. Perimeter guard duty , night (Im painting a picture) 3 hours in ~midnight. Prior to the ground invasion This is about as featureless a landscape as can be imagined, aside from treading water in the open sea. Looking at the sky, you can imagine the number of planes in the sky, if not, there were a hell of a lot of them. I had night vision goggles which made them much more visible. Being that isolated with 0 light pollution makes the night sky spectacular. On one occasion I saw something higher and faster than any of the planes I could see without the NVG's. Struck, I think I'm seeing a UFO, a personal dream BTW. Now I WANT to believe that it is, let me make that much clear. I'm enough of a skeptic even at 21 (then) to settle down. Then I see another and a third. It hadn't immediately dawned on me that satellites should or would be visible to the naked eye, given that isolation. Subsequent observations with and without the goggles on, in some cases consecutive nights confirmed it for me. I'm not trying to get pedantic about it but if it is an unmeasured observation in a dynamic environment without controls, the overwhelming odds say that it's an error in the observation of what actually happened or anomalous which makes it statistically negligible in either case. Its not limiting science but it is defining it within the confines of what can be verified (if there has to be a limit I will concede that one but only because if it cannot be verified read:verifiable then it's not of use) . Anything else, I believe, becomes a twisted synonym for science and is a leap backwards.