Yes, I use 0.001, but the problem is when you want study the effect in the potencial with 1 week of every training, or if you want add sublevels or levels and see the value of this changes.
I don't see problem. You can study difference in potential anyway. It always will be the same, no matter if You use -4 or -5 offset in formula.
For example the total potencial could be lower than the potencial of add levels.
I don't understand what You mean. I scaled only formula discovered by Josef so every problem which You see in new formula exists in old formula as well ;-)
And please note that little inaccuracies in factors always will exist.
This could be confused for average users, the same with negative potencial.
Because we have 4 levels of potential for free. And when potential score is under zero it means that player doesn't exploit true potential, he still uses free potential.