I'd be bloody impressed - assuming of course I was monitoring the test and making sure it wasn't rigged. Also it has to be said that getting something right say 9 times out of 10 would scientically be judged as "lucky" and insignificant. Myself I don't think it would ever happen under controlled conditions. I once was convinced it would be easy - the differences were that huge. I put my convictions to the test. The differences disappeared. They have not come back.