I'd like to add to this thread probability by its very nature is estimating and guessing, attempting to predict a future event, hopefully based upon some stable system that is creating the outcome (either defined or undefined).Golda wrote:Ok we're getting somewhere now. Actually the difference between improbable and probable from a purely statistical/mathematical viewpoint is simply applying an inverse to one in order to get the other.
Earlier in the thread the question went around, what is improbable? This is simply a line in the sand - on one side probable and on the other improbable. In Statistics we call this line the alpha value. As Shazam wrote some times the alpha value can be quite small, (less than 1 outcome in 10,000 events or 0.01%). The most common alpha value used is 1 outcome in 20 events, or 5%. This means that if an outcome happens more than 1 time in 20 events we consider it probable. If and outcome occurs once in every 21 or more events we consider it improbable.
This approach is used in a wide variety of decision-making techniques. It is at the core of statistical process control. It is used for hypothesis testing like T- tests and Z-tests. And it forms a stable basis that independent observers can reach consistent and similar conclusions.
Interesting that Thomas Jefferson framed our alpha value for our criminal courts. "It is better that 100 guilty men go free, than 1 innocent man should be wrongly convicted." What is TJ's alpha value?
