Does the brand matter more than the installer?

Why is this question usually answered without evidence?

Because the two are measured in different units. Ask a contractor whether the brand matters and you will be told the install matters more. Ask a manufacturer and you will be told to buy the good equipment.

Neither answer normally arrives with a number attached, and the reason is that the two things are measured in different units — one is a nameplate rating, the other is a workmanship problem — so they are rarely put side by side.

Source fact: DOE’s Residential HVAC Installation Practices: A Review of Research Findings (June 2018), a systematic review of 44 field-study reports, reports measured performance at 59-84% of rated SEER.

The same review finds at least one fault present in 70-90% of systems, rising to 90-100% once duct systems are included, with undercharge in 29-78% and low airflow in more than 50%.

Source fact: EPA’s ENERGY STAR certified product file publishes SEER2, HSPF2, EER2, heating capacity at 47°F, 17°F and 5°F, and coefficient of performance at 5°F for every certified model, all measured to DOE Appendix M1.

HyreHVAC calculation: Deduplicated to distinct outdoor units and rebrand families — the method published in full on the heat pump rankings — 281,975 certified combinations resolve to 8,465 families, and each capacity band has a measurable efficiency spread from its bottom decile to its top.

HYRE HVAC Research Desk analysis: That gives two ranges in one unit. What follows is the comparison, at three installation-quality assumptions rather than one, and it does not land where the folk wisdom expects on every measure.

What does a brand name actually tell you?

Less than it seems. Source fact: In EPA’s ENERGY STAR certified heat pump file, 1,394 outdoor-unit model numbers are listed under more than one brand name.

6 model numbers appear under 17 separate brands: D5CUHAH18AAK is certified as Airquest, Arcoaire, Bryant, Carrier, Comfortmaker, Day & Night and 11 others.

HyreHVAC calculation: 16.5% of the 8,465 deduplicated families in the file, absorbing 4,051 duplicate listings.

HYRE HVAC Research Desk analysis: For a buyer comparing two quotes, that has a concrete consequence: the choice between two brand names is sometimes not a choice between two machines.

Where a model number is shared, the certified rating is by definition identical, so brand preference in that comparison is buying a badge, a warranty document and a dealer network — all of which may be worth something, none of which is performance.

The limit of that finding, stated: It applies to 16.5% of families, not to all of them; 7,071 model numbers in the file carry exactly one brand.

And ENERGY STAR records no manufacturer field to join on — icm_outdoor_unit_manufacturer is populated on 2 of 281,975 rows — so where two different model numbers are in fact one platform under two badges, this data cannot show it.

The 16.5% is therefore a floor on rebranding, not an estimate of it.

What this changes about the question: "Brand versus installer" is really "specification versus installation", because the specification is a model-level property that a brand name only sometimes predicts.

That is why the rest of this page compares certified models to installation quality, and why the useful thing to write on a quote is a model number, not a manufacturer.

Which spread is wider, the models or the installation?

The installation range is wider. HyreHVAC calculation, and the assumption it rests on: DOE’s 59-84% of rated SEER is a realization ratio: measured seasonal cooling efficiency divided by the rating on the label.

It is dimensionless, so it multiplies whatever the machine is rated at. Expressed the way an equipment spread is expressed — best over worst — it is a range of 42.4%.

The equipment spread: Within a single capacity band, comparing only models a household could actually put in the same place, the bottom decile to top decile of certified SEER2 runs from 3.6% (ducted, <1 ton) to 40.5% (ductless, <1 ton).

In the mainstream 2 to 3 ton ducted band, across 919 models, it is 28.3%: SEER2 15.2 at the tenth percentile, 17.5 at the median, 19.5 at the ninetieth.

The comparison: 42.4% against 28.3%. In all 11 bands measured, the installation range is the wider of the two — by a factor of 1.50 in the mainstream ducted band. The folk wisdom survives contact with the evidence on this measure.

The most conservative version of the same point: If the full 59-84% of rated SEER range feels like a wide net, take one single fault instead.

DOE puts the efficiency benefit of duct sealing alone at 16-41%, and duct leakage is found in 90-100% of systems.

Even the bottom of that one-fault range, 16%, consumes more than half the entire 28.3% brand-to-brand spread — and the top of it exceeds the whole spread on its own.

When does the equipment choice matter more?

In cold weather. A page that only confirmed the trade’s own slogan would not be worth publishing. The same dataset contains a measure where the ordering reverses, and it is the measure that matters most in a cold climate.

HyreHVAC calculation: Capacity retention — heating capacity at 5°F as a share of the certified 47°F figure, both published per model by ENERGY STAR — is what decides whether the system still heats the house on the coldest morning or hands the job to a backup element.

Among ducted certified models it runs from 49.1% at the tenth percentile to 97.1% at the ninetieth — so the top decile holds 98% more of its rated capacity than the bottom, 2.3 times the 42.4% installation range. Across all certified models the same gap is 76%.

HYRE HVAC Research Desk analysis: The asymmetry has a physical reason rather than a statistical one.

Installation faults are subtractive: an undercharged system, a leaking duct or a throttled return all remove performance the machine was capable of. They cannot add capability.

Low-ambient capacity is capability — compressor envelope, injection, control strategy — and if it was never certified into the unit, no amount of commissioning produces it.

2 of the 7 like-for-like measures we compared sit on the equipment side of the line, and both of them are capacity retention at 5°F.

Where that leaves the reader: In a hot or mixed climate, where the heating season is short and the cooling season decides the bill, the efficiency comparison above is the one that applies and the installer is the decision.

In a cold climate, both apply: the model has to be capable, and then the installation decides how much of that capability arrives. The cold-climate rankings hold the certified retention and 5°F COP figures per model, and the underlying study holds the distribution.