In Savannah, Georgia, a homeowner wanted to swap her aging gas furnace for a heat pump. Her existing system pumped 89,000 BTU per hour into a 2,600-square-foot house. An independent building science consultant ran an ACCA Manual J load calculation, the code-required method for determining how much heating and cooling a house actually needs, and got a number that made her old furnace look absurd: 60,000 BTU per hour. That furnace had been fifty percent larger than the house required from the day it was installed.
It gets worse. When the consultant tracked runtime data after installing a two-stage heat pump rated for 24,000 to 37,000 BTU per hour, the system ran mostly on its low stage through the coldest nights of winter. It never once triggered auxiliary heat. Real-world demand was closer to 30,000 BTU per hour, which means the original furnace had been three times the size it needed to be, blasting short bursts of heat at a house that barely needed a whisper, cycling on and off like a jackhammer solving a finishing nail problem, per building scientist Allison Bailes of Energy Vanguard, who documented the case in detail.
That house is not unusual, and it is far closer to the industry norm than any contractor would willingly admit.
How Big Is the Problem
A Florida Solar Energy Center study surveyed over 400 homes and found that more than half had cooling systems sized beyond 120 percent of what a Manual J calculation would specify. A Pacific Northwest study by Lucas (1993) found two-thirds of 75 sites exceeded the Manual J recommendation. An EPA-funded field investigation visited buildings on peak cooling days and discovered that 60 percent had systems 60 percent larger than needed.
When ACCA, the industry's own trade association, analyzed 40 new-construction Manual J calculations, the average result was 1,431 square feet per ton of cooling. Contractors using rules of thumb typically size at 500 to 600 square feet per ton. That means the standard guess oversizes a system by a factor of roughly 2.5 compared to what the engineering would call for, and the contractor who bothers to run the calculation is the exception, not the rule. Green Building Advisor's Martin Holladay put it bluntly in a widely cited analysis: "Most HVAC contractors don't know how to do heating load and cooling load calculations."
So why does nearly every contractor oversize anyway?
Because it works for them. An oversized system still cools the house. Nobody calls back to complain that the air conditioning is too powerful. And the bigger unit carries a bigger price tag. In that same FSEC survey, contractors said the quiet part out loud. "Oversize by 50% so customers will not complain." And: "I go to next half ton up all the time." And the frank assessment from competitors watching the market: "We have observed most other AC contractors and architects grossly oversized units. Customers still believe bigger is better."
What Oversizing Actually Costs You
Short cycling. That is the central mechanical consequence and almost nobody outside the industry has heard the term. When a system has twice the capacity the house needs, it blasts cold air for a few minutes, hits the thermostat setpoint, and shuts off. Five minutes later it fires again. Repeat that forty or fifty times a day instead of the ten to fifteen long, steady runs a properly sized system would make.
Each cold start wastes energy on compressor inrush current, the electrical spike required to spin the compressor up from dead stop. NREL research found that for systems with meaningful parasitic power losses (standby draw from controls, crankcase heaters, and fan motors), a 50-percent-oversized air conditioner carries roughly a 10 percent energy penalty. Henderson (1992) and Lucas (1993) corroborated similar figures independently.
But energy waste is only part of the damage. In humid climates, the real cost is moisture. A properly sized system runs long enough to pull water out of the air, condensing it on the evaporator coil during sustained operation. A short-cycling system never runs long enough for meaningful dehumidification, leaving the house at target temperature but clammy, that uncomfortable feeling homeowners describe as "the air feels wet even though the thermostat says 72." In places like Houston, Miami, or New Orleans, this is not a minor annoyance. It is a mold risk.
Equipment life suffers too. Every compressor start-stop cycle generates mechanical stress that continuous operation does not. A system that cycles fifty times daily instead of fifteen accumulates wear at more than triple the rate, and when the compressor fails at year eight instead of year fifteen, nobody connects it to the sizing decision made a decade earlier.
The National Bill
EIA's 2020 Residential Energy Consumption Survey reports that 87 percent of U.S. households, approximately 113 million homes, use air conditioning, spending an average of $295 per year on cooling electricity. If half those systems are oversized by 50 percent or more, consistent with the FSEC and Pacific Northwest data, and the energy penalty is 10 percent per NREL and Henderson, the national excess cooling cost reaches $1.67 billion annually: 56.5 million oversized systems multiplied by $29.50 per year in wasted electricity each.
That number is conservative. It counts only the electricity penalty from air conditioning. It ignores heating, where the Savannah case demonstrated three-fold oversizing and where HVAC spending nationally averages $1,051 per household per year per EIA. It ignores shortened equipment lifespans, mold remediation costs, and the comfort deficit that drives homeowners to set thermostats lower than they otherwise would, creating additional unmeasured load. A full accounting, if one existed, would plausibly reach $3 to $5 billion annually.
Methodology: 130.2 million U.S. households (Census Bureau 2023) × 87% AC penetration (EIA RECS 2020) = 113.3 million. Half oversized 50%+ = 56.6 million. $295 average AC expenditure × 10% penalty = $29.50 per household. 56.6M × $29.50 = $1.67 billion. Conservative: does not include heating oversizing, equipment replacement acceleration, or humidity-related damage.
Why the Heat Pump Transition Makes Everything Worse
Millions of American households are switching from gas furnaces and conventional air conditioners to heat pumps, driven by Inflation Reduction Act tax credits worth up to $8,000 per installation and by state-level incentive programs. Many of those new heat pumps are variable-speed, inverter-driven systems that modulate their output continuously rather than cycling between full-blast and off. When properly sized, these systems run at low capacity most of the day, ramping up only during peak demand, maintaining steady temperatures with exceptional efficiency and near-perfect dehumidification.
Oversize a variable-speed heat pump, and you negate the entire advantage. Allison Bailes explained the mechanics on Green Building Advisor: put a 6,000 BTU per hour mini-split in a room with 1,500 BTU per hour of cooling load and the unit bottoms out at its minimum output nearly 100 percent of the time. It cannot modulate below its floor. It short-cycles exactly the way a single-stage system would, producing the same comfort problems, the same humidity failures, and the same compressor stress, except now the homeowner paid $2,000 to $5,000 more for variable-speed technology whose entire value proposition requires correct sizing to function.
A 2024 Nature Communications study of 1,023 residential heat pumps found approximately 10 percent were oversized and 17 percent of air-source units fell short of European efficiency standards, with a two-to-three-fold performance gap between the best and worst installations. When the hardware is expensive and the benefits depend on sizing precision, getting it wrong costs more than it ever did with a $3,000 window unit.
A Tool That Fixes This Already Exists
Conduit Tech, now a ServiceTitan company, makes a LiDAR-powered iPad application that performs ACCA-certified Manual J load calculations in 15 minutes. The technician walks through the house with an iPad Pro. LiDAR scans each room, capturing wall dimensions, window sizes, ceiling heights, and door locations as a 3D model. The software applies the building's insulation characteristics, orientation, and local climate data to generate room-by-room heating and cooling loads that meet ACCA certification standards.
Fifteen minutes is the entire gap between guessing and knowing, between a contractor who eyeballs square footage and one who delivers an engineering report.
An HVAC Today reviewer who had spent years with traditional Manual J software (Wrightsoft) tested Conduit against his established tool: "I refused to adopt it until it could match the accuracy of Wrightsoft's Manual J program. Eventually, after countless tweaks, the software hit that mark. In May of 2025, I did my first scan with Conduit Tech, and I haven't looked back since." The 1,800-square-foot home he scanned took 15 minutes start to finish.
Conduit's published adoption metrics tell an instructive story about what happens when contractors actually use data. Companies using the platform report a 32 percent increase in average ticket size, a 21 percent higher close rate, and 126 percent growth in total sales revenue. Wait. Larger tickets with right-sized equipment? The explanation matters: contractors armed with 3D visualizations, branded engineering reports, and AR renderings of equipment placement sell higher-efficiency, higher-margin systems with confidence. Homeowners who see their house scanned in 3D and receive a room-by-room engineering report trust the recommendation, and trust converts to premium equipment choices rather than lowest-bid shopping.
Right-sizing does not require selling less; it requires selling better, with data instead of guesswork and engineering instead of habit.
The Strongest Case for Keeping Things the Way They Are
Defenders of moderate oversizing make a reasonable point about variable-speed equipment: a system oversized by 20 to 30 percent can modulate down, avoiding short cycling while providing reserve capacity for recovery from thermostat setbacks and for those three or four extreme weather days per year when loads spike beyond design conditions. Some manufacturers explicitly recommend selecting equipment above the Manual J calculation to ensure adequate capacity at extreme temperatures, particularly for heat pumps operating in cold climates where output degrades as outdoor temperatures drop.
That argument holds for modest oversizing of modern variable-speed equipment in climates with wide temperature swings. It collapses when applied to the actual sizing practices documented in the field. Nobody needs 20 to 30 percent of headroom when the installed system is 100 to 200 percent larger than the load calculation demands. Nobody is protecting against a once-per-decade cold snap when the furnace could heat two houses simultaneously. And the argument applies only to variable-speed systems; the majority of residential HVAC equipment sold in the U.S. remains single-stage or two-stage, meaning any oversizing translates directly to short cycling with no mitigation.
What You Should Do
If you are building a house: Require your HVAC subcontractor to produce a room-by-room Manual J calculation before specifying equipment. This is already required by the IRC (Section M1401.3 of the 2012 code and all subsequent editions), but enforcement is spotty and many jurisdictions accept a one-page form with a tonnage number and no supporting math. Ask to see the inputs. Check whether the window U-values and insulation R-values in the calculation match your actual specifications. If the HVAC bid comes back with a system tonnage that matches the round-number rule of thumb for your square footage, that is a signal the calculation was reverse-engineered to justify a predetermined answer.
If you are replacing an existing system: Pay $150 to $300 for an independent Manual J calculation from an energy auditor or a RESNET-certified rater before accepting your contractor's sizing recommendation. That calculation will almost certainly show your house needs less cooling than you think, especially if you have added insulation, replaced windows, or air-sealed since the original system was installed. A smaller, higher-efficiency system at the same price point will deliver better comfort and lower utility bills for its entire service life.
If you are an HVAC contractor: Consider what Conduit Tech's data shows. The contractors using LiDAR-based load calculations are not selling smaller systems for less money. They are closing more deals at higher margins because homeowners trust engineering over salesmanship. In a market where every competitor offers three prices on a kitchen table, being the company that scans the house and shows the math is a differentiator that does not require a bigger compressor.
What we did not verify: Conduit Tech's reported metrics (32% ticket increase, 21% close rate improvement, 126% sales growth) are company-published figures, not independently audited. We could not determine what percentage of U.S. HVAC installations currently involve a proper Manual J calculation versus rule-of-thumb sizing; no national dataset tracks this. The FSEC study dates to 1997; while Bailes' more recent data and contractor surveys suggest the problem persists, we have no peer-reviewed national survey of current residential oversizing rates. Our $1.67 billion national cost estimate uses a uniform 10% energy penalty, but real-world penalties vary by climate, system type, and degree of oversizing.← Back to all articles