Coal really falls into two categories:
1. Inferior quality coal that doesn't meet international specifications or coal that can't physically be exported. The cost is basically just the cost of mining it and getting it to the power station.
2. Coal that's of export quality and able to be physically exported. International factors determine what it's worth - the war is one but it was rising even before that.
Same basic situation with gas.
Explaining the issue with the NEM and trading versus physical dispatch, a simplified example as follows.
Suppose that we have 3 fossil fuel power stations, one hydro station and some random wind and solar. Obviously there are more in practice but this is just to illustrate.
Power station "A" uses coal from its own mine that's cheap and plentiful.
Power station "B" uses coal of export grade and the power station itself is near a port. It's more costly to operate.
Power station "C" uses diesel and costs a fortune to run.
Power station "D" is hydro.
The wind and solar work when it blows and shines.
So how should we use this? Well if the aim is to keep the cost as low as possible then it's pretty straightforward:
Run the wind and solar first.
Run "A" next because it's cheap.
Run "B" after that because it's medium price.
Run "C" as a last resort only when everything else is running and isn't sufficient.
Schedule "D", the hydro, based on water availability. It should always be run in preference to "C" for economic reasons but if there's enough water available then we can run it in preference to "B" as well. If we get a flood well then run it in preference to "A" too.
So the basic logic is really quite simple. Run the cheapest that's sufficient at any given time bearing in mind that maintenance outages, breakdowns etc will of course put constraints around this.
Now the basic problem with the NEM is that if A, B, C and D all have different owners, and with their individual traders all working on their own basis, then it does in practice lead to inefficient physical outcomes. It ends up with "A" running at half capacity and "C" running full blast and things like that simply because the traders are trying to make maximise revenue. The best financial outcome for the owners of "A" isn't necessarily with maximum output but rather, it's with the best combination of output times price. If that means "B" and "C" end up running unnecessarily then so be it.
That physical dispatch inefficiency leads to two basic problems:
1. It raises market prices by an amount greater than it raises profit, since very real costs are being incurred.
2. It's a truly terrible way to manage physical resources when it ends up with a solar farm being turned down and diesel being burned purely in an effort to make a profit.
That exists against a backdrop of political and economic ideology and the associated legislation and enforcement thereof.
Now if an engineer or similar person were just left to do what they saw fit, well they'd just run the wind / solar then A, B, C in that order with the hydro placed according to the water level in the dam and expected inflows looking ahead. When it comes to the money, well they'd just point out that this is the cheapest way overall so perhaps the owners of the various stations should just do a deal to split the benefits among themselves and consumers such that everyone wins. Simple.....
Those with a more ideological view, and those who enforce the law, would want nothing of that since it's against the ideology about free market competition, not colluding and so on.
Obviously there's more than 4 power stations in reality but the overall concept still applies. Traders working for the owners of low cost generation price a portion of the output highly, leading to their production being cut, higher cost generation running and price being pushed up. To be fair, they've little choice when they're faced with having to recover all business costs, including fixed costs, from that revenue - they can't simply price at marginal cost otherwise they'll go broke.
So a lot comes down to an ideological view of how the world ought to be there. Technical efficiency versus free markets and so on.
Personally I think they'd be better separated. We need a market design that doesn't lead to inefficient technical outcomes, and the waste of real physical resources, as a result of financial trading and so on.