The first step to discovering whether an idea that seems good is actually good is to review it at another date, which necessitates writing it down.

I was writing them down for a while until I realized the ones I forgot about tended to have obvious flaws almost like my subconscious was filtering things.

Most of my ideas that I've almost-forgot (passed from easy remembrance, only regained with much mental agony) and forgot-but-wrote-down don't have obvious flaws in them. Maybe you have a good mental filter, but for many of us we just have bad memory.

It actually does :)

It doesn't need to be a good idea. It needs to be the next step in the branch; consider being in a conversation when some interruption happens. Interruption resolves, you say "and now back to...what was I saying?" In that moment, the value or proposition might have only been an argument or segue towards a broader point, or a sub-digression, but without that connective node, the entire pattern is lost.

You could counter with "then it wasn't a good idea if you didn't remember it" but it's the structure of how we think rather than the quality of the outputs. Why do we retrace our steps when we misplace something? What does that have to do with anything? It's the scaffolding structure around which we build a mental model. It's so you don't run outside and start looking under rocks. If you believe you wouldn't do that, you have to consider then why, and that is innately tied to the pattern-finding structure of our brain.

Creatives may cling to a lost node as the Rosetta stone, perhaps assigning more weight than is qualified, but then you won't actually know unless you find and explore that branch of thought. Its value lies in its unquantifiability.

Okay, but an idea can never make it from seeming to be to actually being a good idea unless you remember it for a sufficiently long duration.