Why the game is not really about supply and demand

In my experience "supply and demand" have a huge impact on the game. For example, when I sell steel the price goes down and it becomes cheaper for other players to buy the steel they need for the next upgrade. In practice, they buy steel from me instead of a neighboring colony. If i didn't sell my steel those other players would have to make more money to be able to afford it.

Keeping up with what other players are doing definitely matters. If too many players start selling the same resource its price (and profitability) is going to plummet. When multiple players are entering the same market, whoever sells first will do so at a higher price so it is important to pay attention to when other players start making the same resources you do. When a good is scarce, hitting the competition with BM items will allow only you to profit from the high price etc. 

A problem I've found which may explain your experience is that the AI underestimates the profitability of life support resources, which means that the player can safely commit to food, water and oxygen production without risking flooding the market, because the AI (brokenly) is incapable of matching the demand of those goods.

If you're playing at a skill level where the game feels easy enough to build entirely self sustaining, try increasing the difficulty, dropping map size, or even dropping the resources.

As is, a straight up "random" game, imo, has too many resources for a challenge. There's usually plenty of resources for everyone to have several claims on each.

I cannot help but disagree with this post overall, especially because beta 4 has cut the productivity of buildings significantly because of the adjacency bonuses. Perhaps if you was refering to single player I woudl agree with you, but in multiplayer you have to check what is being produced and what you expect the trends are going to be in order to make the most money. Chems usualy crash when everyones level 5, food crashes at level 3 if there's scientifics in the game, electronics become unprofitable if a robot can secure the resources quickly.

Thign is, if a player is being self-sustaining, if they have enough water to cover the costs for making food, then they are playing the game sub-optimally, or perhaps we should say they're taking a risk. Because if water crashes then suddenly they have worthless water tiles. This can be seen easily with aluminium late-game, since aluminium is almost purely an upgrade resource. If you go the other way and crash food then you force a food transition, not as painful but they will still have a large excess of water afterwards. They basically inflate their own water price which is a vulnerability that can be exploited. Alternatively, you can use black market to shut down their water and buy up a ton of it. This artificial price increase can be devestating to someone not paying attention as their farms start producing a huge loss. Then when the water comes back online, you sell your stockpile for a profit since they drove the price of water up. All of this is not talking about the elephant in the room either. Your keeping the cost of water down by supplying it to youself, but is it worth spending 3 tiles on water to make sure it stays at $30, when instead you could spend 3 tiles makign a profit somewhere else? And when your farms become unprofitable, will there be an opening to transition those into something better instead? Sometimes there isn't, and you made the correct play spending tiles on water, but sometimes there isn't. You have to be able to read the market to know what you can and cannot get away with, because once those tiles are spent they're spent forever.

However, if everyone is playing perfectly, then no matter what happens, all the markets will crash (unless map generation is evil). When that happens, offworlds are the way to go. The goal then is to get into offworlds both as fast as possible yet striking the balance of being as cheap as possible and yet also not leaving yourself open to attack whislt you sink your monkey into somethign that takes a while to pay itself off. So, like all RTS's speed is of the utmost essence. Sure, you can be self-sufficient, but that's not how you make the most short-term profit, and in economics short-term profit leads to large term gains. Perhaps you simply haven't been in chalenging circumstances to notice this, but the person who correctly predicts the market and plays with it, instead of being self-sufficient, will gain a small short-term lead which leads to a long term lead.

Now that I read this, I don't even think I answered your actual arguements, but whatever I'm leaving up this bit of rambling anyway

I find that (in multiplayer), it is more important to get a pile of cash to buy other players (versus single player against AI), and to do that one of the best ways is to make sure you are self-sustaining for most of the resources you will need (at least late-game)... otherwise it becomes too much to manage and you end up frantically building and rebuilding, killing your cash pile.

Up to a few minutes in though, it definitely matters. Although I would say it matters a lot more to try to not make yourself the target of storage and EMP attacks!