Does 1 °C lower on the thermostat save 10%? Six cities checked

Turn the thermostat down one degree and save about 10% on heating [5]. It is one of the most repeated energy tips, often credited to the Energy Saving Trust [5]. We checked it against ten years of daily weather in six cities. The tip holds up as a rough guide. But the real figure runs from 8% to 15% [1], and it moves the opposite way to what most people guess.

Why one degree is worth anything at all

A house loses heat roughly in proportion to the difference between inside and outside. Engineers add that difference up over a year as heating degree-days. Each day counts the degrees by which the outdoor mean falls below a base temperature, and heating energy is close to proportional to the total [3]. In the UK the base is conventionally 15.5 °C rather than room temperature [4]. Sunshine, people and appliances supply the last few degrees for free.

Turning the thermostat down 1 °C lowers that base by about a degree. Every day below the base then counts one degree fewer [3]. So the saving is the number of heating days, divided by the year's degree-days [1].

What ten years of weather say

We took daily mean temperatures for 2015 to 2024 from the Open-Meteo historical archive [2]. We counted degree-days at a 15.5 °C base and again at 14.5 °C [1].

The pattern is clear. Mild, long heating seasons save the most per degree: London, Manchester, Edinburgh and Madrid all save around 14% [1]. Hard winters save the least. A Toronto day at −10 °C needs 25 degrees of heat, and one degree less barely dents it [1].

A house that needs heat up to a higher base gets a smaller percentage. That could be a draughty house, or one kept warmer. At an 18.5 °C base, the UK cities fall to 10.6–12.0% [1]. So the 10% tip is a fair, slightly cautious figure for Britain [1] [5].

Nerd corner

The saving is the change in degree-days for a one-degree change in base. That equals the number of days below the base: each one loses exactly one degree [1]. Dividing by the total gives 1 / (base − average outdoor temperature on heating days). That is the formula in the box [1].

A second method agrees: counting degree-days directly at both bases gives 14.4% for London, and the days-below-base shortcut gives 15.0% [1]. They differ because days just under the base drop out entirely.

The model leaves things out. Real heating runs on a timer, a building stores heat in its walls, and a boiler's efficiency changes with load. Treat the box as a good estimate, not a meter reading.

Check yourself

Why does a cold city save a smaller percentage? Its heating covers a bigger temperature gap. One degree is a smaller part of 25 degrees than of 7 [1].

Does that mean a cold city saves less money? Not necessarily: Toronto has 2,896 degree-days a year against London's 1,790 [1]. It heats far more, so a smaller share of a bigger bill can still be more pounds [1].

How low is too low? Public health advice in England is to heat the rooms you use most to at least 18 °C. This matters most for older people and people with health conditions [6].

A notebook entry for energywise.site, written with AI and checked by hand. Every saving is computed from ten years of daily temperatures and cross-checked two ways.

Sources

1. Our own calculation (2026-10-30): heating degree-days at 15.5 and 14.5 °C base from 2015-2024 daily temperatures in six cities
2. Open-Meteo historical weather API (daily mean temperature, 2015-2024)
3. Heating degree day (definition; heating energy proportional to degree-days)
4. University of Strathclyde ESRU: degree days and the UK 15.5 °C base
5. Grimsby Telegraph: the one-degree rule, citing the Energy Saving Trust
6. Healthwatch Kingston (2026): UKHSA keep warm guidance, at least 18 °C

*Corrected 2026-10-30: Toronto's average heating-day temperature is 3.3 °C, not 3.4; and a comparison with a published 1985–2004 degree-day total is removed, because the page it cited gives no such figure.*