I'm sure they do exist, but I need to see harder evidence of these skills. It's not enough that the person is correct in some predictions, they have to also ma

I'm sure they do exist, but I need to see harder evidence of these skills. It's not enough that the person is correct in some predictions, they have to also make few mistakes. And the predictions must be objective and not vague or horoscope-like. Would you be impressed if I made a prediction of a coin flip and was right? What about if I predict a 50/50 choice someone is about to make? What if I have a book of instances where I made correct predictions? Not impressive unless the mistakes are there too.