An insurer cannot predict whether a particular policyholder will make a claim. It can predict with reasonable accuracy how many claims a large group will produce.
Aggregation converts uncertainty into predictability
Individual outcomes are close to random. Across a large enough number of independent policies, the proportion that produce claims settles near a stable figure.
This stability is what makes the business possible. Premiums can be set to cover expected claims plus costs, with a margin for the variation that remains.
The larger and more independent the pool, the narrower that variation becomes, which is why insurers seek scale and geographic spread rather than concentration.
Independence is the fragile assumption
The mathematics relies on policies not failing together. A house fire in one town tells you nothing about a house in another, and the pooling works.
Catastrophes break that assumption by damaging thousands of insured properties at once. The aggregate becomes as volatile as a single large policy.
Insurers manage this by limiting exposure in any one area and by buying reinsurance, which is itself a mechanism for pooling across a still wider base.
Rating factors sort the pool into groups
Since individuals cannot be assessed precisely, insurers group them by observable characteristics that correlate with claim frequency or severity.
Each group is priced against its own expected experience, which is why two policyholders with identical requested cover can be quoted very different premiums.
The factors used are constrained by regulation, and permitted variables differ by jurisdiction and by line of business.
Adverse selection pushes back on pricing
People generally know more about their own risk than the insurer does, and those who expect to claim are more inclined to buy cover.
If pricing does not distinguish between higher and lower risk, the lower-risk group finds the premium poor value and leaves, which raises the average of those remaining.
Left unchecked, the process pushes premiums up repeatedly. Underwriting questions, waiting periods and mandatory participation all exist to limit it.
Why individual experience still affects the price
Claims history is used because it carries information the applicant cannot easily misstate and because it correlates with future claims.
No-claims discounts and experience rating adjust an individual's premium towards their own record while keeping them inside the pool.
The balance between group pricing and individual experience is the central design question in every line of insurance, and it moves as more data becomes available.