HSPF2 Rating Meaning: How the 2023 Test Changed the Number

The same machine scores lower under the newer test because it assumes a harder duct system, so an old figure and a new one cannot be compared — and neither describes a cold snap.

What HSPF2 Is Measuring

HSPF stands for heating seasonal performance factor. It is the total heat a
machine delivers over a heating season, in BTU, divided by the total electrical
energy it consumed doing so, in watt hours. Because the two units are different
it is not a percentage and not a plain ratio, which is why the figure looks
like an eight or a nine rather than something recognisable.

The “2” means the number was produced under the newer test procedure that
took effect in 2023 rather than the older one. Same idea, different test
conditions, and the change is large enough that comparing a figure from one
against a figure from the other is meaningless.

Why your familiar number went down

The central change is the assumed external static pressure — the resistance
the test assumes the indoor blower has to work against. The older procedure was
generous about it, closer to a laboratory duct system than a real house. The
newer one assumes a higher, more realistic resistance, so the blower consumes
more of the electricity in the denominator, and the resulting figure lands
lower.

Nothing about the machine changed. The same equipment tested both ways
returns a smaller number under the new procedure, and a rating that would once
have read as high now reads as ordinary. This has caused a great deal of
confusion for people comparing a quote against equipment they researched a few
years earlier, or against an installed system’s paperwork.

What the number still cannot tell you

Seasonal ratings are averages over an assumed climate, and that is the
limitation that matters most in New York. The season the test models is not the
season an Adirondack house experiences, and two machines with the same HSPF2
can behave very differently at the temperature where you actually need them.

• It is an average, not a floor. A high seasonal figure can
  belong to a machine that is efficient in the mild shoulder months and falls
  away steeply in a cold snap. The figures that describe that behaviour are
  capacity and efficiency stated at a low outdoor temperature, and they are
  published separately.
• It says nothing about backup heat. A system that leans on
  electric resistance backup on cold nights is consuming electricity at a
  coefficient close to one, and how often that happens depends on sizing and
  controls rather than on the rating. That trade-off is the whole subject of
  keeping a fuel-burning appliance as backup.
• It assumes the installation is not fighting the machine.
  The new procedure assumes a higher duct resistance than the old one, but a duct
  system worse than that assumption pushes the real result below the label
  anyway. The
  airflow the system actually achieves is the difference
  between the rating and your house.

The figure to ask for instead, in a New York winter

A seasonal average is a shopping metric. The design-condition figures are the
engineering ones, and they are what a contractor should be quoting when the
question is whether a machine suits an upstate house:

• Rated heating capacity at a low outdoor temperature, in
  BTU per hour, at whatever design temperature applies to your location. This is
  the number that has to meet the building’s heat loss, and it is well below the
  nominal tonnage.
• Coefficient of performance at that temperature. Unlike
  HSPF2 this is a plain ratio — heat delivered divided by electricity consumed,
  in the same units — so a value of three means three units of heat per unit of
  electricity at that moment.
• The minimum output. An
  inverter-driven machine has a floor as well as a ceiling,
  and equipment that cannot turn down far enough short-cycles in mild weather and
  loses the efficiency the rating promised.

Those three, checked against a room-by-room heat loss calculation, answer the
question the seasonal rating only gestures at. Getting that calculation done
properly is what
a load calculation is actually buying.

Turning the rating into money, carefully

The seasonal figure divided by 3.412 converts it into an average
coefficient of performance across the modelled season — a machine returning
that many units of heat per unit of electricity, on average. That conversion is
the honest way to compare a heat pump against a combustion appliance, because
it puts both on a heat-delivered-per-unit-of-fuel footing.

What it cannot do on its own is give you an operating cost, because that
depends on what your electricity and your current fuel actually cost, and those
are local, tariff-dependent and change. Run the comparison with your own
current figures rather than a national average, which is the method set out in
the piece on
the operating-cost arithmetic between a heat pump and a
furnace.

Where the rating stops mattering

Two systems separated by a small difference in HSPF2 will be separated by far
less in practice than either will be separated from its own label by a bad
installation. Ductwork that cannot deliver the airflow, refrigerant charge that
was never weighed in, a machine oversized so badly that it spends the shoulder
seasons cycling, controls set to fall back on resistance heat too eagerly —
each of these costs more than the gap between two respectable ratings.

The rating also says nothing about how long the equipment lasts, which is a
question of maintenance, siting and how hard the machine is worked rather than
of efficiency, and is covered separately in the notes on
how long a heat pump lasts in the New York climate. Nor
does it describe cold-weather behaviour you will actually notice, such as how
the machine handles a January cold snap, which is
a settings and controls question.

Use HSPF2 for what it is good for: a like-for-like screen between candidate
machines tested the same way. Then move to the design-condition numbers, and
spend the remaining attention on the installation, because that is where the
difference between a label and a heating bill is decided.

Frequently Asked Questions

What does HSPF2 mean?

Heating seasonal performance factor, measured under the test procedure that took effect in 2023. It is the total heat a machine delivers across a modelled heating season, in BTU, divided by the total electricity it consumed, in watt hours. Because the numerator and denominator are in different units the result is not a percentage and not a plain ratio, which is why it looks like a figure around eight or nine.

Why is HSPF2 lower than the HSPF I remember?

Mainly because the newer procedure assumes the indoor blower is working against a higher duct resistance, which is closer to a real installed system than the older test’s assumption. More of the electricity in the denominator goes to moving air, so the number comes out lower. The same machine tested both ways returns two different figures, and nothing about the equipment changed.

Can I compare an HSPF2 figure with an HSPF figure?

No, and doing so is the most common mistake with these ratings. They come from different test conditions, so the comparison is meaningless in either direction. Compare HSPF2 with HSPF2. The same applies to the cooling metric, where SEER2 replaced SEER for the same reason and cannot be set against the older figure.

Is a higher HSPF2 always the better machine?

It is a better average over a modelled season, which is not the same thing as better where you live. A high seasonal figure can belong to equipment that is efficient in mild weather and falls away steeply in a cold snap. For a New York winter the figures that decide it are the rated capacity and the coefficient of performance at a low outdoor temperature, which are published separately.

How do I turn HSPF2 into a coefficient of performance?

Divide it by 3.412, which is the number of BTU in a watt hour. That gives an average coefficient of performance across the modelled season — heat delivered per unit of electricity consumed, in consistent units. It is the honest basis for comparing a heat pump against a combustion appliance, though it still does not give you an operating cost without your own local energy prices.

Does the rating account for backup resistance heat?

Not in a way you can rely on for your own house. How much a system leans on electric resistance backup depends on how it was sized, how the controls are configured and how cold it gets where you are, and resistance heat runs at a coefficient close to one. A machine with an excellent rating that calls for backup often will disappoint on the bill.

Does a good HSPF2 guarantee a low heating bill?

No, because the installation can cost more than the difference between two respectable ratings. Ductwork that cannot deliver the required airflow, refrigerant charge that was never weighed in, equipment oversized enough to short-cycle through the shoulder seasons, or controls that fall back to resistance heat too eagerly will each swamp the label.

Where do I find the HSPF2 for a specific model?

In the manufacturer’s published performance data for the exact indoor and outdoor combination, because the rating belongs to the pair rather than to the outdoor unit alone. Ask the contractor to quote the specific matched system rather than a model family, and ask for the low-temperature capacity and coefficient of performance from the same document while you are there.