Chris,

Roger's method is as clear as daylight. If you take one prediction at a time and find out whether it fit the parameters, it either does or it doesn't. Then you have the other elements, such as was it significantly greater than chance or no better than random chance. It's pretty straight-forward.

I have every confidence that any evaluations he has made in regard to my work are totally correct. And I've learned a great deal from him about where I've made my mistakes and in time I think I'll do much better.

It is really difficult for anyone who doesn't predict earthquakes to have any understanding of the degree of difficulty. It looks quite easy and it is everything but. There are so many considerations that if you weigh each one it would be difficult to distinguish which element would be more important than the other because an error in any one of them causes a loss.

Don's an over achiever and tries to create the shortest window available and make something fit into it. He would do so much better if he gave himself a little slack; but that's not his way, so he is making what he does either superior or a zero. I don't like to lose that much, so I cut myself a little slack. I admire him for his dedication to excellence. I can't quite do that myself.

Anyway, prediction is an evolution, you either evolve and do better or resolve that you have much to learn and a long way to go.

Humble...humble.

Petra