Science behind "Stability"

[svscorp]

I am curious, what is used behind the "Stability" in forecast, and the advice to keep it below 25%?

[nbrown02]

I took the same logic from Troy Magennis' work:

[svscorp]

So stability, it's the "error" (variation/deviation) then? Usually high stability is good, if the aim for this stability to keep it lower, then considering it as deviation/variation makes sense. Or I might still misunderstand the point.

[nbrown02]

Yup - I could/should probably change it actually to be something clearer (reverse the calculation and do "aim for 75%+") for example