Some solutions rely on convoluted chains of logic that are strictly dependent on every single statement being true. They are more likely to have hidden “divide by zero” problems that may be easily noticeable to the experienced practitioner but are invisible to the layman. Simple solutions might have errors too, but they will be much more obvious. Also, complicated chains of logic “feel” correct because a lot of the steps will be verifiably true; people sometimes forget that all of the steps have to be true for the entire argument to have any truth.
Complicated stories seem more likely to be logically sound to a midwit, but simple strategies are actually far less likely to have hidden land mines.
I’m a big fan of looking for simple explanations for problems, or at least starting with simple explanations and testing them before overcomplicating your rationale. The midwit memes seem like unrealistic parody, but you see these situations in the wild all the time. Self-proclaimed “experts” desperately want to have complex solutions to problems with complex systems — hell, one very well may be required. But at least invalidate the Pareto-fitting, more-likely simple solutions first.