Right, that is the purpose of insurance, to take risk and spread it across a population.

Now the simplest way of doing that is you decide whether someone is "insurable" or "uninsurable" and then for everyone insurable, you define payout criteria and a fair pay in rate (premiums) which is based on your ability to calculate their risk and taking some extra on top for providing the service.

Your skill at:

1. assessing risk correctly as to whether you take them on as clients

2. calculating their risk correctly and mapping it to a price to charge them (premiums)

3. defining payouts in a way that allows you to pay out when things happen to your clients so others trust you to pay out, but not so often that you have no working capital

broadly determine how well you'll do.

You can do all kinds of other complicated things on top of that, but from what I can tell, the fundamental idea seems to be that given those considerations, the insurer pays out, so the fact that someone has a high risk home should be priced into their premiums or they should not have been taken on in the first place.

Now you appear to dislike that people who have different risk profiles are grouped together, what I'm trying to understand is how that works.

For example, in the case of the house burning down:

1. The insurer pays the homeowner out and increases their premiums

2. The insurer pays the homeowner out and places them into a different risk category of people who own similar homes, but have had their house burn down, works out their new premiums, which are now likely much higher as they're in a riskier category and it's likely that population is smaller.

I assume you're arguing for something like 2 to happen?

Or is it something else?