The short version: based on how much air your ducts leak into space you aren't living in, we calculate wasted energy. Then, based on your energy rates, we determine your potential savings. The range reflects how leaky your ducts turn out to be — the low end is a fairly tight never-tested system, the high end a leaky one — with a thorough aerosol seal assumed in every case. The expected case reproduces what field studies measured in real homes.

The equation is the measured-leakage method from Connecticut's 2026 Program Savings Document duct-sealing study — the same algorithm the state's efficiency programs use to pay for this work. The methodology is not CT-specific; it applies anywhere, and we run it on New York and Connecticut climate and rate data.

• Leakage amount — 20–36 CFM25 per 100 sq ft (expected 27) for ductwork that has never been leak-tested, which is most ductwork, including much of what was installed recently. Anchored on Lawrence Berkeley National Lab's measured median of 27 for existing single-family homes and a Vermont field sample at 28.6. No study links leakage to the age of the house or the ductwork — three-year-old systems on Long Island averaged 450 CFM25, the same as 60-year-old ones — so we don't ask when it was installed. Ductwork that has been tested and passed to code is at 4 CFM25 per 100 sq ft or under, and these savings don't apply to it.
• How much of that leaves the house — this is the number that decides everything. Field measurements put leakage to outside at 0–20% of total leakage for ducts in a basement connected to the house (Synertech, Hudson Valley homes) and 54–85% for attic, crawlspace and garage ducts (Long Island, Raleigh, Puget Sound, ORNL). We use 50% for ducts in an attic, garage or unfinished basement (the blend of those two), 80% for a cooling-only attic system, and 10% inside the envelope (wall and floor cavities aren't perfectly sealed from outside, which is why the 2021 IECC and New York's 2025 code now test these ducts too).
• How much of that costs you money — the study's Duct Loss Factor by location: 0.70 for attic ducts, 0.50 for an unfinished basement, blended to 0.60 for the attic-garage-basement option, 0.25 inside the conditioned envelope (the study keeps a flat 25% loss on return-side leaks regardless of where they are).
• Blower electricity — leaks anywhere make the air handler run longer to satisfy the thermostat. We estimate annual fan energy as airflow × 0.35 W per CFM × run hours, and credit sealing with the share of operating airflow it recovers, discounted by half because leaks inside the house still deliver some heat. It's small — typically 20–60 kWh a year — but for ducts entirely inside the envelope it's most of the bill savings there is.
• How much sealing removes — we assume aerosol (Aeroseal-type) sealing, which is pumped through the ducts from the inside and reaches leaks no one can see — and we assume the job is done well, removing 90% of the leakage to outside, which is what Raleigh's Aeroseal field study measured. Hand sealing with mastic only reaches what the contractor can get to and typically removes 50–65% (ORNL's randomized 80-home test: 49–50%; Raleigh: 56–59%), so expect roughly two-thirds of these numbers from a hand-sealed job. Connecticut's program also discounts modeled savings by a 0.65 realization rate to cover jobs that underperform; we don't apply it, because we're assuming the seal is done right.
• How hard your system works — equivalent full-load hours for the NOAA station nearest you (Bridgeport, Danbury or Poughkeepsie), from the same study.
• How efficiently it heats and cools — from your equipment's age. Over 20 years: 72% AFUE gas/propane, 75% oil, SEER2 8.5, HSPF2 6.2. 10–20 years: 80%, 83%, SEER2 11.0, HSPF2 7.0. Under 10: 92%, 87%, SEER2 14.5, HSPF2 8.2. Cooling carries the study's 0.75 discretionary-use adjustment. More than one heat source splits the heating load evenly. Heat-pump savings are modeled only — no field study has measured them in the Northeast.
• Cross-check against the field. With the attic factors, a 2,200 sq ft never-tested gas home comes out at about 103 therms a year; with the basement factors, 18. The best field measurements say 107 (ORNL, ducts outside the envelope) and 8–23 (Connecticut's billing analysis; ORNL, basement-only). The model reproduces the measurements, which is why we trust it. The blended option lands between the two; where your ducts actually run is exactly what the test settles.
• Rates — your ZIP sets your electric and gas utility. These are marginal rates — what you actually save per unit you don't use — not bill averages that include fixed charges. Con Edison electricity 36.8¢/kWh from an actual Westchester bill (June–July 2026); Con Edison gas $1.50/therm marginal from the October 2026 tariff leaf (the $2.50–3.00 figures you see elsewhere are bills divided by therms, inflated by the fixed charge). Heating oil is between October 2026 spot ($6.30 NY, $6.15 CT) and EIA's winter-season forecast ($5.26). Oil and propane aren't utility-regulated, so those are regional.
• What it costs — we test, we don't seal, so we don't quote a price here. For reference, aerosol sealing runs about $2,300 per system in this market and hand-sealing with mastic about $1,200. Fairfield County: EnergizeCT's Home Energy Solutions visit includes a duct test and basic sealing for a $75 co-pay, and the Advanced Duct Sealing rebate pays $1,750 toward the first system and $1,000 toward a second ($600 / $350 for cooling-only) through December 31, 2026. New York has no duct-sealing rebate; the federal 25C credit ended December 31, 2025.

What we deliberately don't claim: you'll see "duct sealing saves 20–30% of your energy bill" all over the internet. ENERGY STAR's actual statement is that 20–30% of the air moving through the ducts is lost to leaks — which is not the same as 20–30% off your bill, and ENERGY STAR attaches no bill-savings figure to it at all.

Sources: CT 2026 PSD duct sealing study · CT HES billing analysis (R1983) · ORNL 80-home field test · Synertech Hudson Valley basement study · Long Island central AC study · LBNL-6378 · Raleigh hand vs aerosol sealing · EnergizeCT Advanced Duct Sealing · ENERGY STAR duct sealing