Storms cost a Florida pool route 1.5 workdays a year. The real cost is one lost week every 12–20 years.Storms cost aFlorida poolroute 1.5workdays a year.The real cost isone lost weekevery 12–20years.Storms cost a Floridapool route 1.5 workdaysa year. The real cost isone lost week every12–20 years.

Method: wind at the county, days lost at the route

We asked a narrow question: in a normal year, how many route workdays does storm wind take from a pool service business, and how lumpy is that loss? We answer it in three steps, all in the scripts that come with this report.

1. Where NOAA’s wind radii put storm-force winds

NOAA’s National Hurricane Center publishes HURDAT2, the official “best track” for every Atlantic storm: position, peak wind and, since 2004, estimates of how far tropical-storm-force (34 kt, 39 mph), 50-kt (58 mph) and hurricane-force (64 kt, 74 mph) winds extend in each quadrant. We interpolated every track to 10-minute steps for the 22 seasons from 2004 to 2025. For each county we placed a point at its 2020 center of population (Census), because that is where the pools are. If the point fell inside a storm’s wind radius for its quadrant, the county got an event at that wind class; each storm counts once per county, at its strongest class. Post-tropical stages count too (Sandy was post-tropical at landfall); a tropical-only version is in the data.

2. Rates for a route’s area

A route covers an area, not a point, and 22 seasons is short for rare events. So each county’s rate pools the event counts of every county whose population center lies within 40 miles, weighted by population: it is the rate at a typical customer’s home in that area. The chance of at least one hurricane-force event in N years is 1 − e^{−rate × N}. That formula assumes storms arrive independently; in reality they cluster (2004–05, 2017, 2022–24), so we also report what actually happened in rolling ten-year windows.

3. From wind to lost workdays and revenue (a model)

Expected days lost per year = Σ (events per year in a wind class × days lost per event in that class). Revenue on hold = days lost × accounts × monthly price × 12 ÷ 260. Whether that revenue is really lost depends on your billing policy: if you credit missed visits it is lost; if you bill a flat monthly rate it shows up as catch-up work instead. The days per event are assumptions, anchored on the evidence in the table.

Cross-check: across the 1,132 study counties with a FEMA value, our expected days lost rank almost the same as FEMA’s National Risk Index hurricane frequency (Spearman ρ = 0.93) and its expected-annual-loss score (ρ = 0.86). Both rest on NOAA’s track data, so this checks our processing, not the underlying record.

Sample and coverage

Florida carries the most storm days by far

Expected route workdays lost per year (central case). The darkest counties sit on Florida’s Atlantic coast and around Lake Okeechobee, on the Outer Banks of North Carolina, and on the Louisiana coast.

Each county weighted by its 2020 population, so inland counties with few people count little.

Source: Ruberio model; population: FEMA National Risk Index county table (Census-based), Dec 2025.

Source: Ruberio model on NOAA HURDAT2 (2004–2025) wind radii; counties pooled within 40 miles. Census 2023 boundaries.

Where in Florida a hurricane-force hit is most likely

The map shows the modeled chance that hurricane-force winds reach a typical home in a route’s area at least once in the next ten years, at the 2004–2025 rate (the upper end of our range); at that rate 78% of Floridians live where it is 50% or higher. The Space Coast, the Treasure Coast and the counties around Lake Okeechobee are highest, followed by Southwest Florida; the Jacksonville area (Bradford, Baker, Clay) had no hurricane-force winds at its population centers in 2004–2025 and the inland north is lowest.

Source: Ruberio model on NOAA HURDAT2 (2004–2025). Assumes storms arrive independently at the 2004–2025 rate.

Small every year, big once

For a route owner the average is not the problem: $720 a year on a 70-account route is about 0.6% of the year’s revenue. The problem is that the loss comes in one block, in the same weeks that power, roads and customers are also out. JPMorgan Chase Institute’s bank data showed that in the worst week of Hurricane Irma, most small firms in the Miami area took in at least 82% less than in the same week a year earlier; most balances recovered within two to three weeks.

Average year = 1.49 workdays (low–high 0.74–2.98). One hurricane hit = 6 workdays; bad hit = 12 (both assumptions).

Share of 9,407 county events in the study area by month the winds arrived, 2004–2025 (no events in other months).

Source: Ruberio analysis of NOAA HURDAT2.

Two busy decades: why the ten-year chance is a range

A 22-season window is long enough to see the pattern, but it was not an average stretch. Counting how often a hurricane’s center passed within 50 nautical miles of a county’s population center, Florida saw 3.0× as many passages per county per season in 2004–2025 as in 1975–2003. North Carolina went the other way.

Hurricane-strength center within 50 nautical miles of a county’s 2020 population center; per county per season ×1,000.

Source: Ruberio analysis of NOAA HURDAT2 (1975–2025).

Limitations

• Wind radii are estimates of maximum extent. NHC gives the farthest reach of each wind class per quadrant, mostly over water. Inland, sustained winds are usually lower, so our event counts lean high, especially for tropical-storm-force winds.
• Days lost per event are assumptions. No public data measure missed pool visits after storms. We anchored them on power-restoration records and business surveys; the low and high cases show how much they matter. Prep days for storms that miss are not modeled, which leans low.
• Wind is not the only hazard. Flooding, storm surge and long outages can keep a route down longer than wind alone suggests (Ian in Lee County, Michael in the Panhandle). Rainfall and surge are not in this model.
• Pooling smooths local extremes. Averaging over 40 miles steadies the rates but dilutes landfall counties: Bay County, where Michael came ashore in 2018, comes out lower than its own record.
• Storms cluster. The ten-year probabilities assume independent storms; actual decades were lumpier, which is why the rolling-window figure is lower.
• Post-tropical storms lift the Northeast. Counting post-tropical stages (Sandy, for example) raises New Jersey and New England; the tropical-only version in the data lowers them.
• Prices and route sizes are illustrative. They come from consumer cost guides and operator-submitted data, not a survey.
• Upside is not modeled. Storm recovery work is often billed separately and can offset part of the loss.
• The 2026 season is missing. HURDAT2 is finalized after each season; this year’s storms are not included.

What an owner can do with this: a continuity plan

Find your county in section 05 and your route size in section 06, then fill in the worksheet. The first checklist is for before June 1; the other two are for storm week and the days after.

Before the season (by June 1)

• Write down your missed-visit policy (credit, catch-up, or recovery fee) and put it in your service agreement.
• Set aside the reserve from the worksheet; know which credit line you would use.
• Photograph each customer’s equipment pad and note pump, filter and heater models (useful for insurance and for triage).
• Store chemicals dry, off the floor and above flood level, and keep acids, liquid chlorine, cal-hypo and trichlor apart from each other; keep the SDS sheets with your records.
• Read your policies. Business interruption cover typically needs direct physical damage, has a 48–72-hour waiting period and excludes flood without a separate flood policy (Triple-I); many small operators have none, so check your business owner’s policy. Check that your commercial auto policy covers a flooded truck (comprehensive).
• Decide your crew policy: who works, who is paid, and when it is safe to drive the route again.
• Save your fuel, chemical-supplier and parts-supplier contacts offline; agree on a backup supply source.
• Prepare two customer messages: “before the storm” and “after the storm” (what you will check, what costs extra).

When a storm is forecast

• Tell customers what you will and will not do. Industry guidance: don’t drain pools if you don’t know the groundwater level (an empty pool can lift out of the ground); turn off pool equipment at the breaker (PHTA via ICC).
• Move the route schedule forward rather than skipping it; fuel trucks and fill chemical stock early.
• Secure the truck, chemicals and records; back up your customer list and schedule off-site.

After the storm

• Don’t start cleanup until the area is safe (UF/IFAS Extension). Don’t enter a pool or touch equipment until power to it is confirmed off, and don’t restart submerged motors or heaters until they are dry and checked by an electrician.
• Triage: equipment underwater or damaged first, then green pools, then routine stops. Tell customers about any extra fees before you do the work.
• Keep receipts and photos. Check IRS disaster relief (deadline postponements by county), SBA disaster loans (physical damage up to $2 million; EIDL at up to 4% for up to 30 years, but SBA does not count a loss of expected profits alone as economic injury) and Florida’s Small Business Emergency Bridge Loan when it is activated.
• Afterwards, compare the days you actually lost with this model and adjust your reserve for next season.

Sources

1. NOAA National Hurricane Center, Atlantic hurricane database HURDAT2, 1851–2025 (file hurdat2-1851-2025-092326.txt). https://www.nhc.noaa.gov/data/hurdat/ — accessed Oct 30, 2026.
2. Landsea, C. W. and J. L. Franklin (2013), Atlantic hurricane database uncertainty and presentation of a new database format, Monthly Weather Review 141, 3576–3592 (HURDAT2 format and wind radii).
3. FEMA, National Risk Index, county table, version December 2025, via FEMA’s public ArcGIS service “National_Risk_Index_Counties”. https://hazards.fema.gov/nri/ — accessed Oct 30, 2026.
4. U.S. Census Bureau, 2020 Centers of Population by County (CenPop2020_Mean_CO). https://www2.census.gov/geo/docs/reference/cenpop2020/county/ — accessed Oct 30, 2026.
5. U.S. Census Bureau, Cartographic Boundary Files 2023, counties and states 1:20,000,000. https://www2.census.gov/geo/tiger/GENZ2023/shp/ — accessed Oct 30, 2026.
6. NOAA NHC, Tropical cyclone climatology (season peak Sep 10). https://www.nhc.noaa.gov/climo/ — accessed Oct 30, 2026.
7. U.S. Energy Information Administration, Today in Energy: Hurricane Irma (Sep 20, 2017), Hurricane Michael (Oct 22, 2018), Hurricane Ida (Sep 15, 2021). https://www.eia.gov/todayinenergy/ — accessed Oct 30, 2026.
8. American Public Power Association, Hurricane Ian outage update (Oct 5, 2022) and Hurricane Milton restoration (Oct 10 and 13, 2024). https://www.publicpower.org/ — accessed Oct 30, 2026.
9. Florida Municipal Electric Association, Hurricane Idalia restoration (Sep 4, 2023). https://www.flpublicpower.com/ — accessed Oct 30, 2026.
10. WMNF, “In Florida, most power has been restored after Debby” (Aug 8, 2024); Suncoast News, Isaias restoration (Oct 14, 2026). Accessed Oct 30, 2026.
11. Federal Reserve Bank of Dallas, Texas Business Outlook Surveys special questions on Hurricane Harvey (Sep 25, 2017). https://www.dallasfed.org/news/releases/2017/nr170925.aspx — accessed Oct 30, 2026.
12. Florida Gulf Coast University, Regional Economic Research Institute, Lee County Business Climate Survey, Q4 2022 (Dec 1, 2022). https://www.fgcu.edu/cob/reri/news/reports/lee-ebcs-fourth-quarter-2022-report — accessed Oct 30, 2026.
13. JPMorgan Chase Institute, “Bend, Don’t Break: Small Business Financial Resilience After Hurricanes Harvey and Irma” (Feb 2018). https://www.jpmorganchase.com/institute/all-topics/business-growth-and-entrepreneurship/report-bend-dont-break — accessed Oct 30, 2026.
14. Pool & Hot Tub Alliance Recreational Water Quality Committee, “Treating flooded pools and spas,” reprinted by the International Code Council (Jun 1, 2020). https://www.iccsafe.org/building-safety-journal/bsj-technical/treating-flooded-pools-and-spas/ — accessed Oct 30, 2026.
15. UF/IFAS Extension Indian River County, “Pools and hurricanes” (May 14, 2024, updated Apr 2025). https://blogs.ifas.ufl.edu/indianriverco/2024/05/14/pools-and-hurricanes/ — accessed Oct 30, 2026.
16. U.S. Small Business Administration, Disaster assistance: physical damage loans and Economic Injury Disaster Loans. https://www.sba.gov/funding-programs/disaster-assistance — accessed Oct 30, 2026.
17. Internal Revenue Service, tax relief for taxpayers affected by Hurricane Isaias in Florida (FL-2026-07, Oct 14, 2026). https://www.irs.gov/newsroom/ — accessed Oct 30, 2026.
18. Triple-I (Insurance Information Institute), “Do I need business interruption insurance?” https://www.iii.org/article/do-i-need-business-interruption-insurance — accessed Oct 30, 2026.
19. Ready.gov, Business preparedness. https://www.ready.gov/business — accessed Oct 30, 2026.
20. Pricing inputs (consumer cost guides and crowdsourced operator data, not a census): HomeGnome, pool cleaning cost (May 12, 2026); PoolDial, pool service pricing benchmarks 2026 (Jun 2026) and pricing guide (Feb 2026); SmartService, pool service route (Jun 15, 2026). Accessed Oct 30, 2026.

About this research