spread of many runs

The same model run hundreds of times with the uncertain inputs redrawn each time, and all the outcomes shown at once as a band with a few sample paths inside it. The band answers a different question than any single run. Not what happens, but how wide the range of what happens is. It is the one sub-mode that shows the bad tail, which is often the number the decision turns on.

The same model run hundreds of times with the uncertain inputs redrawn each time, and all the outcomes shown at once as a band with a few sample paths inside it. The band answers a different question than any single run. Not what happens, but how wide the range of what happens is. It is the one sub-mode that shows the bad tail, which is often the number the decision turns on.

Example

Best when chance drives the outcome and planning on the average would hide the tail that matters.