One degree on the thermostat, and the physics behind it

Loss is proportional to the difference, not the setting

In steady conditions, the rate at which heat escapes through a wall, a roof or a window is proportional to the temperature difference across it. Not to the indoor temperature, and not to the outdoor one, but to the gap between them. That single relationship explains most of what heating controls do.

Work through an illustrative case. If the house is held at twenty-one degrees while it is five degrees outside, the difference driving the loss is sixteen degrees. Drop the setpoint to twenty and the difference becomes fifteen, which is a reduction of about six per cent in the rate of fabric loss.

Now change the weather rather than the setting. On a mild day at twelve degrees outside, the same one-degree reduction takes the difference from nine to eight, which is a saving of over eleven per cent. The identical adjustment is worth roughly twice as much in mild weather, because it is a larger fraction of a smaller quantity.

Why the annual saving is smaller than the instantaneous one

That six or eleven per cent applies to the rate of heat loss at a moment, and the annual heating bill is not the same quantity. Several things dilute it.

Free heat gains — from occupants, cooking, appliances, lighting and sunlight through windows — supply part of the heating requirement, and they do not change when the setpoint does. In a mild shoulder season those gains may cover a substantial share of the demand, which means the heating system runs less and the saving applies to a smaller base.

The heating system’s own efficiency varies with how it is being run, and not always in the direction you would expect. A condensing gas boiler recovers more energy when its return water is cool, which happens when the system is running gently rather than flat out, so a lower setpoint can help twice. A heat pump behaves similarly and more strongly, since its efficiency depends directly on the flow temperature it has to reach.

The old argument about leaving the heating on

The claim that it is cheaper to leave heating on constantly at a low level than to let a house cool and reheat it comes up perpetually, and it deserves a careful answer rather than a dismissal.

The physics is unambiguous on the main point. A cooler building loses heat more slowly, so a house allowed to drop while empty loses less total energy over that period than one held at temperature. Reheating costs energy, but only enough to restore what was lost, which is less than was saved. Setback wins on that comparison.

The complications are real, though, and they are why the argument persists. A building with high thermal mass responds slowly, so a short setback may achieve little before reheating begins. A heat pump reheats most efficiently at a low, steady flow temperature, and forcing a rapid recovery pushes it to run harder and less efficiently, which can erode or reverse the saving. And in a poorly insulated house, letting internal surfaces cool below the dew point can encourage condensation. The general rule favours setback; the exceptions are specific and they are not imaginary.

The thermostat is measuring one place

A single thermostat controls the whole system from one point in one room, and that room is not representative. It is usually a hallway, often near a door, sometimes above a radiator, occasionally in direct sun for part of the day. The temperature it reports is the temperature there.

This is why a house can feel cold at a setting that ought to be comfortable, and why the fix is often positioning rather than a higher number. A thermostat in a draughty hall calls for heat the rest of the building does not need. One in a warm sunny room shuts the system down while bedrooms stay cold.

Zoning, thermostatic radiator valves and room-level controls address this by allowing different spaces to be held at different temperatures, and the saving comes not from running cooler everywhere but from not heating rooms nobody is in. That is a larger effect than a degree on the main setpoint in most houses, and it is often cheaper to implement.

What this does and does not justify

The reasonable conclusion is that lowering the setpoint saves energy in a predictable and calculable way, that the effect is proportionally larger in mild weather, and that it is one of the very few measures available at no capital cost whatsoever.

It is also a comfort decision rather than a technical one, and comfort is not a trivial consideration. Cold homes have real consequences for health, particularly for older people and for anyone with a respiratory condition, and this is emphatically not an argument for being cold to save energy. The point is that the arithmetic is knowable, so the trade can be made deliberately.

Where a setpoint reduction does have a hidden benefit is in what it enables. A house that is comfortable at a lower indoor temperature — because draughts have been eliminated and cold surfaces warmed by insulation — is a house where the reduction costs nothing in comfort at all. Fabric work and thermostat settings are not competing measures. The first is what makes the second painless.