http://www.chartzero.com/ come back and report your score I tried it twice got around .570 both times
I started bad, then learned how they choose their stocks, then my score went up, from a low of 25 to now around 75. The stock choice should be random. In addition even if you and I did well, it could be just the luck of the draw.
Kinda fun, are you supposed to do any particular number of trials? I was doing quite well for a while, and then reverted towards the mean possibly from lack of concentration or bad luck. Ended up with a bit over 0.6 in maybe 30 charts.
The approach is flawed. It does not measure the right things. If one of you is the owner of this, I have ideas to build the right game that would benefit people.
I reverted to 50%. The approach is flawed. A better way than "where will this be in 2 weeks" is "will this first go a) up or b) down x% in the next x days." If you did it like that, then it would be testing something. It's an interesting approach. I think it's definitely an interesting way to potentially improve pattern recognition.
Not only the approach, but the whole premise behind those games is severely flawed. Why make an educated guess where an instrument will be trading two weeks from now? That's throwing away an extremely large amount of information worth mountains of gold. If I'm playing KK HU vs. an opponent with A2s face up who'll take any bet I throw at him, I'll c/c as long as my PE > 50%, and shove when its 100. Why would you play it any other way?
That was my thought, too. Two weeks out is very tricky except when there's a strong trend setup. I did very well on those, but dreadful on all the others.