Let's take a closer look at this one.
"Compute the good with the largest price differential among the locations... destination is the planet with the highest price. "
Given this method, and the player has opened up a newly generated area with a high price for the good in question, and let's put it quite far from the center of the explored/generated areas. You'll have a swarm of traders headed there - since they're all using the same decision making method.
Now, that's only one problem that your implementation will run afoul. There will be dozens, hundreds more as the code invariably grows. Whether or not you can spot them early in the dev cycle and what you have to do to kludge a fix in later on, is another question.
Sorry, I'm not sure I understand the problem. You say that the player places, for example, a factory station that is in a remote system, and sets the station to offer a higher-than-average price for delivery of a certain good, is that correct?
Yes, as you point out, a good number of traders will then recognize and aim to exploit this differential. Not all traders, mind you, because the algorithm I gave only selects a subset of locations to consider, so not all traders will consider the player's factory at once. But yes, you will have a fair number of traders competing to exploit this differential. How is that unacceptable, or different from real life? If Wal-Mart offered to pay people $5 for every empty Coke can they brought in, I'm sure we'd see a pretty substantial rush on Wal-Mart, regardless of how remote it might be!
Again, perhaps I'm not understanding, but it doesn't seem like a problem to me - the behavior is as expected. In fact, it points to a desirable property of the simple algorithm I gave. If you offer to pay significantly higher than market price for a good, yes, you will see an influx of trade. This is great to know if, for example, you need to have goods rushed to your station quickly. You can set a high price to intentionally create this behavior!
These sorts of issues will come up exponentially as the project grows, and discussing one or another method is pointless - which is why I said "clouds of fog". The details of one issue is besides the point. What is relevant, is that you intend to do all of this on your own and with no experience of your own or of others to guide you. Somehow you think you'll manage to do what very few novice programmers can do - keep the "big picture" firmly in their mind whilst coding the minutae. That's worrying.
No, not really, it's not at all pointless. Look, we're finally having an actual discussion, about actual content! You can wave your hands all you want and say things are difficult, but if you can't provide any evidence, how is that any different from what you've accused me of doing?
The details of one issue are not beside the point. We can have a talk about why it's actually not an issue. The issue is that you're generating issues! Which makes it very much productive to discuss, for example, the 'issue' that you just brought up.