Hail Damage on Roof

How the tables on this site were built

Two tables carry this site. This page says where each cell comes from, what the figures assume and what we chose to leave out.

Contents
  1. The question the site answers
  2. The threshold table
  3. The identification table
  4. Evidence we accept
  5. What we left out
  6. What we do not claim
  7. Keeping it current

A threshold quoted without its conditions is easy to misuse. This page sets out how the tables were assembled, so that a reader who cites them can also cite their limits.

The question the site answers

A person looking at marks on a roof needs to decide two things: whether hail made them, and whether hail of that size fell on the property on the date of loss. Every page is organised around one or both.

The threshold table

The hail size thresholds come from three published sources, in this order of preference.

  1. Haag Engineering's 2002 conference paper, which reports ice-ball impact tests on nine roofing products and the firm's field thresholds from surveys after real hailstorms. We read the paper itself and reproduce its figures without rounding.
  2. The threshold table published by Haag's Scott Morrison in 2009, as reproduced in IIBEC's journal in 2013, for clay tile and slate, which the 2002 paper does not cover.
  3. An article by Haag's chief executive in a claims journal, for the rupture threshold of metal panels, which reports the firm's 2010 testing.

A threshold here means the onset of functional damage, defined by the researchers as a loss of water-shedding capability or a reduction in expected service life. The figures assume hard hail, a perpendicular impact, reasonable support under the roofing and a roof in relatively good or midlife condition. The laboratory stones were freezer ice, harder and denser than natural hail. We state these conditions wherever the table appears.

For impact-rated shingles we give no hail-size threshold, because the rating standards do not produce one. We describe the tests instead.

The identification table

The descriptions of hail impacts and of each look-alike come from Haag Engineering's 2004 paper on asphalt shingles and its 2019 inspection protocol, from an article by a Haag engineer in the National Roofing Contractors Association's magazine, and from IBHS and insurer guidance written for homeowners. Where the sources disagree, as they do on granule loss, the text says so and gives each position.

Evidence we accept

  1. Primary research and protocols from engineering firms and research institutes, read in full.
  2. Government sources: NOAA and the National Weather Service for hail sizes, storm reports and radar products, and state statutes for deadlines.
  3. Trade associations and manufacturers for test standards and product bulletins.
  4. Insurers and insurance bodies for claim statistics and policyholder guidance, labelled as theirs.
  5. A vendor's own pages for what its product contains and costs. Skybridge Intelligence's report is described only from the company's live pages, and the limits the company states are reproduced with it.

We do not use forum posts, contractor marketing pages or figures we could not open on the access date.

What we left out

Some pages we wanted could not be read on the access date and are not cited: the Texas Department of Insurance's consumer pages, NOAA's Storm Events database pages, the National Weather Service page on the one-inch hail criterion and the full members-only IBHS reports. Thresholds for synthetic and stone-coated steel roofing, and the specific steel-ball sizes in UL 2218, are omitted for the same reason.

What we do not claim

We have not tested roofing or inspected roofs. Test results are reported from the studies that produced them. Nothing here is legal advice or a statement about what any insurer will pay.

Keeping it current

We re-check every source before each hail season and whenever a reader reports a change. Each page shows its last update and a log of material changes, and each source shows the date we opened it.