I agree with the first part. But in fairness, he did say his calculations assume 5.5 hours of sun a day. One think he did not address is maintenance costs. Nothing lasts forever and once out of warranty, things like a failed inverter or charge controller can be drastically change that ROI. ,
The
potential productive capacity of a PV system has no bearing on the calculation of a system's
actual ROI.
All that matters for actual ROI is the difference in total earnings and savings between the two scenarios of making the investment in a solar PV system and not making the investment.
Which means all you should include in the Return part of an ROI assessment is the value of the energy actually produced and consumed, which has two basic values:
i. the savings value of the grid imports replaced, and
ii. the earnings value of selling the energy to the grid.
In an off-grid scenario the latter is zero, so the ROI is simply the cost of the alternative energy supply you have replaced, i.e. what would have it cost had that energy been purchased from the grid. One should, in fairness, use the cheapest grid supply option in their location for comparison.
The fact a system is capable of producing more than is actually consumed isn't relevant. Unused capacity has no value.
At least some grid-tied systems can get some return for excess capacity sold to the grid. But such a system may cost more as well (in order to meet regulatory approval standards).
Things like earnings from crypto mining have zero bearing on solar PV ROI calculations because you can mine equally well with grid power. The only financial difference between those two scenarios is the cost of the electrical energy used.
As to having power available when the grid is down, that definitely has value. But equally, the grid has value as a backup for when your own solar PV system is down. Outages can happen with the grid supply and with your own supply. Which backup option is then considered more valuable?
And power backup can be provided via other options, e.g. a petrol powered generator, which despite the downsides is still in many cases (but not all) a very effective and low cost way to have emergency power in an outage. Why spend several thousand $ on a solar PV / battery backup when a $500 generator and $20 of fuel can do the same job in many (but not all) instances?
Of course the capex considerations also come into play, e.g. a rural property where to get grid supply it might mean a $100k cost for the poles and wires connection by the local energy distribution company, vs building your own system. That's a different kind of assessment and one where much longer term considerations need to be factored in such as who is responsible for the cost of maintenance, upkeep and replacement of the power supply infrastructure, as well as the land improvement value of having grid access vs not.