I learn from books and others and then I use the knowledge I get in my own way.
Based on what you're looking for (supplement retirement income with low risk, would be happy with a 20% return) and experience (have a trading account and have been trading stocks for years), I think your approach is not unreasonable: learn from various sources and then use your own judgement.
Problem with a 20% low risk return is not that it doesn't exist, but that it's not accessible to retail individuals. Hedge funds that make such figures are closed entities, already having enough money for their profitable systems. I used to work for one that made 50% on options and that only on some $50M of capital. The hedge fund itself had about $1B and had no choice but to place them in stocks, with much lesser results.
And what banks offer as "Asset Management" is a joke. Virtually any bank has such a division, like put the bank name and add "asset management" and you have it. I can see why people will rather trade on their own than lend their money to the "specialists" in the banks, those specialists are in general no better than cats picking stocks (
https://www.forbes.com/sites/freder...ofessionals-at-stock-picking/?sh=4031542e621a ).
The trading strategy that I developed recently seems to be compatible with what you're looking for (high return, low risk or at least known and managed risk). But there's still a few problems with it:
1) I've so far only ran it on backtests.
Detractors and trolls will quickly jump at my throat and dismiss me as ridiculous: only backtests, lol. But youz stupid if you say this because ANY trading strategy starts with theory (strategy outline) and backtests. Backtests are the if and only if condition of any further steps. If it doesn't work on historical data, would you trust a strategy to run live on your own money?
And finding something that DOES work on backtests is excruciatingly difficult because finance is a hugely competitive domain so almost anything you can think about was already tried and arbitraged away. Not the best example since it was mostly used by amateurs but Quantopian (now defunct) had some 20 million strategies designed and tested and still went bankrupt. Think of that when you start your path of high optimism that you'll beat the market in spite of the huge drones of people who failed to beat it before you.
So passing backtests is the first step, also quite depressing because it immediately exposes your "genius guru strategy" that you pick on Reddit or pay money for: you ran it and the numbers mercilessly decimate it.
2) Even if it passes next stage of validation (run live on a virtual account - hedge funds also have this step before going fully in production), the cruel reality is that financial market is a gambling casino and you can't keep beating the casino if you tell the casino how you do it. So the only way to make money is to keep the strategy proprietary and exploit it as long as it works - which in theory could be a long time. But another thing I'm throwing in the face of geniuses who repeat the meme "run it on yer own money". A trading strategy is like any retail product such as a smarthpone. To make money you need to scale: have the phone sell in huge numbers so you make billions of capital for it, have the strategy run on a billion dollars in order to really extract the value in it. If one is already working at a hedge fund that has the capital, they already have the infrastructure for exploiting the strategy (question still remains how much would the researcher of the strategy gets out of it). If one's not having the billions of a hedge fund at their disposal, then the only way to put a strategy to use is to scale by selling multiple copies to retail traders. But the first copy you sell has let the bird out of the box and from then on, there's no telling how much the strategy will hold before it becomes common knowledge and stops working.