No, I'm not missing the point.

Once the window closes, the prediction can't be dropped.

I see no problem with making a change to the prediction to either extend a window or make a change to the magnitude. I'm not worried about location right now since I don't have that down to a knats ass yet, but when that time comes that'll come up. All a change does is cause a recalculation of the probability. Extending the window increases the probability of getting a hit by chance while at the same time increasing the number of hits required to become statistically significant. Increasing the magnitude would lower the probability of getting a hit by chance while lowering the magnitude would increase that probability.

We need to know whether or not a person knows what they are doing by making a change. By not allowing changes keeps the predictor from demonstrating that.

Dennis