Skip to content
SwimSolar
Methodology

How we work out every number

This is the most important page on the site. Everything else is an application of what is on it.

By Scooter M.Published How we work out every number

What we do not do

We have not tested any of the products on this site. We do not own them. There is no lab, there is no test pool, and there is no 60-day trial in anybody’s backyard.

That is stated first because burying it would read as concealment, and because it is unusual: most of the pages competing with ours lead with a testing claim. Four of the eight competitor pages we examined say they handled the products themselves, and two put it in the headline.

We are not going to imitate that, and not out of modesty. Claiming to have run tests we did not run is a fabrication, and for an affiliate publisher it is also an FTC problem — endorsers are required to make truthful, substantiated statements. So instead of a claim you cannot check, this page gives you a method you can.

What we do instead

  1. Compile what manufacturers actually publish, and say plainly when they publish nothing. In this category that happens constantly, and it is a finding rather than a gap in our research.
  2. Apply published physics from the Department of Energy, the EIA, the USGS and NASA — evaporation load, BTU accounting, fuel energy content, water density, solar irradiance.
  3. Compute the answer for your pool, from your gallons, your temperature target and your energy rate, rather than quoting a range that fits nobody.
  4. Show every formula and every assumption on the page, beside the answer rather than in a footnote.
  5. Cite every source with a link that works, and record the date we read it.

The point of all of it is that you can re-run our arithmetic on a phone calculator in about a minute. That is what we have instead of a test lab, and we think it answers your actual question better — you want to know what a cover will do for your pool, not what one did in somebody else’s backyard in Kansas.

The formulas

The whole model, in three lines
BTU to raise a pool by ΔT = gallons x 8.33 x ΔT
Hours to heat = BTU required ÷ (heater BTU/hr x efficiency)
Cost = (BTU required ÷ 1,000,000) x cost per million BTU delivered

Everything else on this site — every cost table, every timing estimate, both calculators, every worked example in an article — resolves to these three lines. They live in one module in the codebase, so the number printed in an article and the number produced by a calculator cannot drift apart.

The constants, and where each came from

Every figure below was read from the linked source on the date shown. None of them is rounded to look tidier, and none is remembered.

Pool energy figures
FigureValueSource
Share of pool heat loss caused by evaporation70%DOE
Heat carried off per lb of 80°F water evaporated1,048 BTUDOE
Heating-cost saving from covering when not in use50-70%DOE
Make-up water reduction from a cover30-50%DOE
Chemical consumption reduction from a cover35-60%DOE
Recommended water temperature, active swimming78-80°FDOE
Recommended water temperature, recreational82-84°FDOE
Cost increase per +1°F of target temperature10-30%DOE
Effect of a 7 mph surface wind on energy use+300%DOE

These are the Department of Energy’s figures. DOE’s own Energy Saver pool pages have been taken down — every URL under energy.gov/energysaver/ that this subject needs returned 404 when we checked on 2026-09-02, including swimming-pool-covers, swimming-pool-heating-and-maintenance and heat-pump-pool-heaters. We cite the page that reproduces and attributes them and that resolves, and we link an archived copy of the DOE original alongside it. We are not going to cite a URL that 404s, and we are not going to pretend the live page still works.

Physical and energy constants
FigureValueSource
Water density at 39.2°F62.424 lb/ft³USGS
US gallons per cubic foot7.48052Definition
Energy to raise 1 gallon by 1°F8.33 BTUDerived; also the trade figure
Energy content of 1 kWh3,412 BTUEIA
Energy content of natural gas1,036 BTU/cu ftEIA
Energy content of propane91,452 BTU/galEIA
Solar energy at top of atmosphere1,360 W/m²NASA
Fraction absorbed at the surfaceabout 48%NASA

The 8.33 figure is derived rather than quoted: USGS gives water's density as 62.424 lb per cubic foot, a cubic foot is 7.48052 US gallons, so a gallon weighs about 8.34 lb cold and about 8.33 lb at pool temperature — and one BTU is by definition the heat needed to raise one pound of water by one degree Fahrenheit. It happens to be the number the pool trade uses too, which is a useful check rather than the reason for it.

Energy prices (US residential averages)
FuelPricePeriod
Electricity18.34 cents/kWhJune 2026
Natural gas$24.09 per thousand cubic feetJune 2026
Propane$2.674 per gallonWeek ending 30 March 2026

All from the US Energy Information Administration. These are national averages and we say so everywhere they are used: residential electricity varies roughly threefold across US states, so a national average is the starting point for arithmetic and never a prediction of your bill. Both calculators let you type your own rate in, which is the only version of this number that is true for you.

The assumptions — and they are assumptions

Three numbers in our model are not published by anybody. They are choices we made, and they are labeled as such everywhere they appear. Both calculators expose all of them as editable inputs, because the honest answer to “how efficient is your heater” is “read the data plate”.

What we assumed, and why
AssumptionValueWhy
Gas heater thermal efficiency82%A federal minimum standard exists at 10 CFR 430.32(k), amended by 88 FR 34624 with compliance required from 30 May 2028. We could not open a page that prints the percentage, so we treat this as our assumption rather than quote a regulatory figure we did not verify.
Heat pump coefficient of performance5Manufacturers publish COP at warm test conditions where it looks best, and it falls as air temperature drops. This is a deliberately mid-range figure for warm-weather operation, not a best case.
Electric resistance COP1Resistance heating converts electricity to heat essentially 1:1. There is nothing to assume here beyond stating it.
Season length used in seasonal examples150 daysA stand-in for a typical northern-US season stretched at both ends. Every seasonal figure on the site says which day count produced it.

How the delivered-cost figures are built

The number that decides every heating argument on this site is the cost of getting a million BTU into the water. It is built in two steps, and the second step is the one the category leaves out.

Step one — what a million BTU of fuel costs to buy
Natural gas: $24.09 per Mcf ÷ 1.036 MMBtu per Mcf= $23.25
Propane: $2.674/gal ÷ 91,452 BTU/gal x 1,000,000= $29.24
Electricity: $0.1834/kWh ÷ 3,412 BTU/kWh x 1,000,000= $53.75

At this stage electricity looks ruinous and natural gas looks cheap, which is exactly the comparison that gets published and exactly why it misleads.

Step two — divide by what the equipment actually delivers
Heat pump: electricity ÷ COP 5.0= $10.75 per million BTU delivered
Natural gas: gas ÷ 82% efficiency= $28.36
Propane: propane ÷ 82% efficiency= $35.66
Electric resistance: electricity ÷ COP 1.0= $53.75
Solar: no fuel= $0.00

The ranking inverts completely once efficiency is applied

A heat pump moves heat rather than making it, delivering roughly five units for every one it draws. That is the whole reason electricity goes from the worst option to the best one, and it cannot be seen without dividing through.

Why the published temperature numbers disagree

This is the piece of analysis we are most confident is original, and it is the reason this site exists. The buying guides in this category tell readers a solar cover gains 4 degrees, or 10 to 15, or “up to 15”, or 5 to 8 per twelve hours — a three- to four-fold spread that nobody reconciles.

The spread exists because three different physical quantities are being reported as one. Daytime solar gain, overnight retention, and cumulative equilibrium are different things with different mechanisms and different magnitudes. Every temperature claim on this site names which of the three it is, states its conditions and cites its source. The full reconciliation is here, including the arithmetic showing why a full pool cannot physically gain 4 degrees in an hour of sun, and what a thermometer that reads it is probably measuring.

How we choose products

The same criteria on every page, and they are the criteria that survive not having tested anything:

  • What is actually published about it, weighted by who published it. A manufacturer figure beats a retail listing, and a retail listing beats nothing.
  • Whether the specification that matters exists at all. A brand that publishes no lifespan figure cannot be compared on lifespan, and we say so rather than filling the gap.
  • Cost per year of expected service, not sticker price.
  • Whether the thing will actually get used. A cover that is too heavy to deploy alone saves nothing, which is why a reel outranks a thicker blanket in our recommendations.
  • We name what to skip. Every roundup on this site contains at least one product we would not buy, and says why.

How we handle prices

Prices come from the live Amazon data layer or they do not appear. There is no second price source anywhere in this site’s code, and no price is ever typed into an article — which makes “we never show you a stale price” a structural fact rather than a promise.

A price older than 48 hours disappears by itself and the button falls back to “check price on Amazon”. If you are reading this and seeing that on every button, it is because the price feed is not live yet on this site — which we would rather tell you than paper over with a number we remembered.

Telling us we are wrong

The whole point of showing the arithmetic is that it can be checked, so please check it. If a formula is wrong, a constant has moved, a cited page has gone dead or a product has changed, info@swimandsolar.com reaches us.

We will verify it, correct the page, and date-stamp the correction where you can see it rather than editing quietly. That commitment is written down on the editorial policy page.

Sources