Glossary
Probable maximum loss
A modeled estimate of how big an insurer's loss from a single severe catastrophe could be, stated at a probability such as a 1-in-100-year event.
Also called: PML, 1-in-100 loss, 100-year PML, 1-in-250 loss, return period loss
Probable maximum loss, or PML, is a catastrophe-model estimate of how large a loss from one severe event could be. It is usually quoted by return period. A 1-in-100-year PML is the loss with about a 1% chance of being met or exceeded in any given year. It does not mean the worst loss imaginable, because bigger events exist, and it does not mean a loss that occurs once a century.
Insurers use PML to decide how much reinsurance to buy, and rating agencies and regulators use it to test whether surplus can survive a major storm. Say an insurer's model puts its 1-in-100 hurricane loss at $2 billion and its surplus at $800 million. It would want reinsurance covering most of the gap above what it can absorb, arranged in layers like the ones in a reinsurance tower. Florida law directs Citizens Property Insurance Corporation to try to buy reinsurance covering its projected 100-year PML.
For a homeowner the figure is indirect but useful. Different models give different PMLs, so a company's number is an estimate, not a fact. An insurer with thin reinsurance against its own PML is more fragile after a big storm. Look at its financial strength rating, and read how catastrophe models feed into reinsurance costs and your premium.
Last reviewed October 10, 2026