The advice from the lucky ones is a tricky thing.
Maybe they did x, y, and z and were lucky. As a storyteller, it is natural to assume that x, y, or z are related to it. This may lead to wrong recommendations.
Think of the person who smokes like cigarettes, drinks like fish, and lives to be 100 years old. It is not clear whether his habits helped him get lucky.
Luck may not be a strategy, but making yourself lucky may be a strategy.
Luck is a strategy. An unpredictable, of course, but if it works, it works. A useful strategy might be: I will build a flexible model and apply information and testing to find effective methods. One of the strategies that support this strategy may be to appear where luck can help me. If I can persist long enough, I will be lucky.
But this is very different from the wrong association between past behavior and lucky results.