Skip to content
hailroofcheck
Menu
Hail science

Hail Size Chart: Inches, Objects and What Size Damages a Roof

Hail sizes from pea to grapefruit in inches, when hail counts as severe, and the hail sizes that have damaged shingles, wood, tile and metal in tests.

Updated November 10, 202614 sources6 min read

The National Weather Service chart below runs from pea-size hail (1/4 inch) to grapefruit (4.5 inches). Hail becomes “severe” at 1 inch, the size of a quarter, and the NOAA Storm Prediction Center calls 2 inches, about a lime, “significant severe.” Older, thinner roofs can start to take damage near the 1-inch line. In storm surveys by Haag Engineering (field thresholds for roofs of all ages, presented at an American Meteorological Society conference in 2002), damage began at 1 inch on 3-tab asphalt shingles, 1.25 inches on laminated and cedar shingles, 1.75 inches on concrete tile and 2.5 inches on gravel-covered built-up roofing. Age, wind and hail hardness move those lines, and so, in lab tests by the Insurance Institute for Business & Home Safety (IBHS), does repeated small hail.

Hail size chart: everyday objects in inches

The National Weather Service recommends this conversion table for turning spotter reports into hail diameters in Storm Data, its storm record. NOAA’s National Severe Storms Laboratory (NSSL) gives the fall speeds, with what it calls “much uncertainty.”

ObjectDiameter (NWS)CategoryExpected fall speed (NSSL)
Pea0.25–0.375 inbelow severe9–25 mph (all hail under 1 in)
Small marble0.50 in (NWS advises against “marble-size”)below severe9–25 mph
Penny0.75 inbelow severe (severe until 2010)9–25 mph
Nickel0.88 inbelow severe9–25 mph
Quarter1.00 insevere25–40 mph (1–1.75 in)
Half dollar1.25 insevere25–40 mph
Walnut or ping pong ball1.50 insevere25–40 mph
Golf ball1.75 insevere25–40 mph
Lime2.00 insignificant severe44–72 mph (2–4 in)
Tennis ball2.50 insignificant severe44–72 mph
Baseball2.75 insignificant severe44–72 mph
Large apple3.00 insignificant severe44–72 mph
Softball4.00 insignificant severe44–72 mph
Grapefruit4.50 insignificant severecan exceed 100 mph above 4 in
Computer CD/DVD4.75–5.00 insignificant severecan exceed 100 mph above 4 in

The objects are rough guides: the NWS notes that a real quarter is 15/16 inch and that “softballs range in size from 3.50 inches to 5.09 inches.” The Storm Data instruction doesn’t recommend “marble-size” or “dime-size,” because marbles vary and a dime is smaller than a penny. The NWS spotter guide says to compare hail to coins or balls, “but not to marbles, or measure the hail with a ruler.”

When hail counts as severe, and why 1 inch

As the SPC’s FAQ puts it, the NWS defines a severe thunderstorm as one “producing hail that is at least 1 inch in diameter or larger,” or gusts of 58 mph or more, or a tornado. The SPC’s significant severe storm produces hail “at least 2 inches in diameter or larger,” gusts of 75 mph or more, or “a tornado that produces EF2 or greater damage.”

The 1-inch line dates from January 5, 2010, when NWS Service Change Notice 09-52 moved every forecast office’s warning criterion from 3/4 inch (penny) to 1 inch (quarter); Central and Western Region offices had already been testing 1 inch. Haag’s lead author sat on the NWS committee that reviewed the criteria; the 2002 paper concluded that “hailstones one inch in diameter begin causing damage to some of the older, thinner roof products.” Storm Data still enters every hail report of 3/4 inch or larger.

What size hail damages a roof, by material

The Haag paper counts as damage “a diminution of water-shedding capability or a reduction in the expected long-term life of the roofing material.” The first column comes from its storm surveys; the second from its lab, which fired ten ice balls of each size straight at each product.

RoofingField threshold, all ages (Haag 2002, Table 3)Lab: first ice-ball size that damaged samples (Table 2)
3-tab asphalt shingles1.00 in1 in on 11-year-old organic-mat shingles (5 of 10 hits); 1.25 in on 11-year-old fiberglass (6 of 10)
Laminated (“30-year”) asphalt shingles1.25 in1.5 in on 11-year-old samples (6 of 10)
Cedar shingles1.25 in1.25 in on 11-year-old samples (3 of 10)
Cedar shakes1.50 in (medium)1.5 in on new heavy shakes (5 of 10)
Fiber-cement tiles1.50 in1.25 in on new tiles (2 of 10)
Concrete tiles1.75 in1.25 in on new flat tiles (2 of 10); 2 in on S-shaped tiles (8 of 10)
Built-up roofing with gravel2.50 in2 in on 8-year-old samples (3 of 10)

Haag’s lab ice balls were frozen tap water, harder and denser than natural hail, which the authors called a “worst case scenario,” and every hit was square-on. Haag reported “good correlation” between its lab and field results, though the two columns still differ by up to half an inch either way.

Tile. In a 2004 tile study, Haag reported that no concrete tile broke from 1-inch ice balls; four of 13 products broke at the corners at 1.25 inches, and 2.5-inch ice broke all of them. Clay tiles held at 1.25 inches, but every corner broke at 1.5. For most tile products tested, the threshold was “about 1.5 in.”

Metal. In a 2013 IBHS demonstration with 1-, 1.5- and 2-inch hailstones, damage to new standing-seam panels “was cosmetic in nature, leaving visible dents but no punctures”; panels laid over a layer of 3-tab shingles dented deeper. IBHS called it a capabilities demonstration. How a policy treats cosmetic dents is a question for your insurer; see hail damage on a metal roof.

Siding. We found no published size threshold for siding. IBHS wrote in 2013 that no standardized hail tests exist for siding, and in 2014 that impact data for non-roofing materials are “more scarce.” NSSL says wind-driven hail “can tear up siding on houses, break windows and blow into houses.” More on hail damage on siding.

Why the same hailstone does different damage

The IBHS claims study of the May 24, 2011, Dallas-Fort Worth storms lists the factors beyond stone size:

  • Age. A Roofing Industry Committee on Weather Issues survey after that storm “found that older roofing materials had reduced resistance to hailstone impacts.” Standard impact tests use new products.
  • Wind. Strong winds mean “the hailstones may be propelled at an angle,” hitting walls, windows and roof slopes that face the storm.
  • Hardness and density. Hailstones range from hard ice to slush; a denser stone of the same size hits with more momentum.
  • Shelter. Trees or taller buildings upwind can shield part of a house.
  • Impact rating. New Class 4 products meet the test criteria when struck by 2-inch balls, but IBHS notes the ratings “do not guarantee that no damage will occur in a hailstorm.” See impact-resistant Class 4 shingles.

Small hail adds up as well. In an IBHS lab study published in 2025, asphalt shingles that weathered for two years and took repeated hits from 0.7- and 1-inch ice were “approximately ten times more susceptible” to damage, measured as granule loss, from later 2-inch hail than new shingles. On average, a single hailstone of 1 inch or smaller can cause nearly 30% of the granule loss of one 2-inch hailstone. The test covered six products.

How to size the hail you saw

IBHS says, “Don’t retrieve hailstones while it is still hailing.” Afterward, measure the widest point of stones from the grass with a caliper or ruler. A photo beside a ruler beats a guess: IBHS says people tend to overestimate hail under 1.75 inches and underestimate hail over 2 inches. Dents on AC fins and soft metal help too, though in Haag’s tests aluminum box vents dented to about half the ice ball’s size and flue vent caps to twice. After the 2003 Dallas-Fort Worth storm, Haag found lead and aluminum items “can be dented by relatively small hail,” but the dents are “permanent and cumulative, and could indicate prior hailstorms.” What those marks look like on shingles is on what hail damage looks like on a roof.

The largest hailstone on record in the U.S.

The national record fell in Vivian, South Dakota, at about 6 p.m. CDT on July 23, 2010. The National Climate Extremes Committee verified a diameter of 8.0 inches and a weight of 1.9375 pounds (1 pound, 15 ounces), both U.S. records. Its 18.625-inch circumference didn’t beat a stone that fell in Aurora, Nebraska, on June 22, 2003.

Hail reports for your area

The SPC keeps NWS hail reports in a database that starts in 1955 and cautions that the data “may not accurately reflect all storm events.” Our hail history tool counts those reports by state and county for 2016–2025. A report means hail of that size was seen nearby, not that your roof was hit.

Common questions

How big is golf ball size hail?

The NWS converts “golf ball” to 1.75 inches. That’s above the 1-inch severe line and above Haag’s field thresholds for asphalt and cedar shingles.

What size is baseball size hail?

It’s 2.75 inches in the NWS table, past the SPC’s 2-inch “significant severe” line.

Can pea-size or penny-size hail damage a roof?

Not from single hits in the tests we found: 3/4-inch ice didn’t damage 11-year-old organic 3-tab shingles in Haag’s lab, and ice of 1 inch or less didn’t damage any tile product tested. Repeated small hail is different: IBHS’s 2025 lab work found that repeated small hail plus weathering made shingles about ten times more susceptible to the next large hail.

What size hail is considered severe?

One inch, about a quarter; before January 5, 2010, the NWS warning criterion was 3/4 inch. Two inches or larger is “significant severe.”

Questions to ask your roofer

  • What size hail fell at my address, and where does that figure come from?
  • Is the damage functional (cracks, punctures, exposed mat) or cosmetic?
  • Do the dents on gutters and vents match this storm, or could they be older?
  • Is my roof covering impact-rated, and can you show me the label or product data?

Not sure whether the storm hurt your roof? Answer a few questions from the ground in the hail damage check.

Sources

  1. National Weather Service — Instruction 10-1605, Storm Data Preparation (July 26, 2021), PDF pp. 45–46: section 19 Hail and Table A3, Hail Conversion Table
  2. National Weather Service — Weather Spotter's Field Guide: What to Report (no update date shown; checked Nov 10, 2026)
  3. NOAA Storm Prediction Center — Frequently Asked Questions, 4.2 and 4.3 (page last modified Nov 10, 2026)
  4. National Weather Service — Service Change Notice 09-52: minimum hail size criterion, effective January 5, 2010 (PDF p. 1)
  5. NOAA Storm Prediction Center — Severe Weather Maps, Graphics, and Data Page (last update Apr 24, 2026)
  6. NOAA National Severe Storms Laboratory — Severe Weather 101: Hail Basics (no update date shown; checked Nov 10, 2026)
  7. National Climatic Data Center, National Climate Extremes Committee — Verification of the Vivian, SD record hailstone (memorandum, Nov 1, 2010), PDF p. 1
  8. Marshall, Herzog, Morrison and Smith, Haag Engineering — Hail Damage Threshold Sizes for Common Roofing Materials, 21st Conference on Severe Local Storms, American Meteorological Society (Aug 2002), PDF pp. 1–4, Tables 1–3
  9. Marshall, Herzog, Morrison and Smith, Haag Engineering — Hail Damage to Tile Roofing, 22nd Conference on Severe Local Storms, American Meteorological Society (Oct 2004), PDF pp. 2, 4, 7
  10. Marshall, Herzog, Morrison, Rae and Mitchell — The Dallas-Ft. Worth, TX Hailstorm: 5 April 2003, 22nd Conference on Severe Local Storms, American Meteorological Society (Oct 2004), PDF pp. 3–4
  11. Brown and Pogorzelski, Insurance Institute for Business & Home Safety — Claims Analysis Study of May 24, 2011 Hailstorms in Dallas-Fort Worth, AMS 94th Annual Meeting (2014), PDF pp. 2–3, 9
  12. Meisenzahl, Giammanco and Hedayati, IBHS — Sub-severe hail: the missing piece in assessing asphalt shingle risk in North America, Frontiers in Materials (published Sep 29, 2025): Abstract, Results and Table 6
  13. Insurance Institute for Business & Home Safety — Hailstorm Demonstration Results (February 2013), PDF pp. 2–4
  14. Insurance Institute for Business & Home Safety — How to Measure a Hailstone (PDF, no date shown; checked Nov 10, 2026)