It is one of the most useful presuppositions I know, and one of the hardest to actually hold: everybody makes the best choice available to them at the time they make it.
Notice what it does not say. It does not say that people make good choices. It does not say that the choice was wise, or kind, or that you should accept its consequences. It says that from inside that person's model of the world, with the resources and the information and the emotional state they had in that moment, that option looked like the best one available.
This matters enormously in practice, and here is why. The moment you decide that someone chose badly out of stupidity or malice, you have stopped being able to influence them. There is nothing left to be curious about. You are now managing a defective person.
But if you assume the choice made sense from where they were standing, an entirely different question opens up: what would have to be true about your world for that to be the best available option? That question is the beginning of every useful conversation I have ever had with anyone about their behaviour.
It applies to you too, and this is the part people find harder. That decision you still cannot forgive yourself for — the job you left, the thing you said, the years you spent in something you should have ended earlier. You did not have the options then that you can see now. You had different information, a different nervous system and a different set of fears.
The point of holding this is not absolution. Consequences remain. Accountability remains. What changes is the quality of your enquiry: instead of asking why am I so stupid, you can ask what was the choice actually available to me, and what would need to be different for a better one to be on the list next time.
That second question has an answer. The first one never does.