With winter approaching, I wonder: Is it better and more economical for the heating system to stop/tone down heating during the night (like letting it go down to 14C) or try to keep a constant temp like 19-20C day and night?

I would guess in 'time of use' there is not much difference between letting it go down and keeping it at the same temp (since the small hits of heat during the night will be pretty much the same as the big hit in the morning.)

So I would like to know... which option is best for the electricity bill, and what is best to also keep my heating system in good shape the longest?

Also, is the same true during summer with AC? I'm feeling more that the 'let it go down and then one big push back up' should be the better option, but my step-dad is adamant its the 'keep a constant temp' approach that is best.

  • Two possible factors, but not generally applicable in most areas for residential electricity are "time of day" rates (charging more for use during the day, less for early morning & evening, even less for nights) and demand charges (a monthly charge based on the highest single usage amount (typically a 30-minute interval). Where those apply, usage at night and/or usage in a way that spreads demand instead of having peaks can save a lot of money. Where those don't apply, it gets more to calculations relating to insulation, thermal capacity, etc. Oct 4, 2021 at 16:14
  • For a standard heating system (not some fancy groundwater heat exchanger or some such) it's (slightly) more efficient to let the temperature dip during the night, since the heat lost through the insulation will be less.
    – Hot Licks
    Oct 4, 2021 at 16:49
  • 1
    Your step-dad is stuck in "urban legends from high school days." Guess what will happen if you ask for him to provide an analysis based on physics & thermodynamics. Oct 4, 2021 at 17:58
  • How well insulated is your house? How efficient is your heater? What is the charge per BTU that you pay to run your system? Run the system one way for a month and note the cost, run it the other way for the next month, note the cost. Look at your bill (or possibly the local weather service) to find out how many "heating days" there were during each of the months and then figure out your cost/heating day. Guaranteed to answer your question for you.
    – FreeMan
    Oct 4, 2021 at 18:05
  • 2
    No experimentation is needed. The whole thing is governed by a physical law: heat movement through insulation is proportional to the difference in temperature of the 2 places. So a 50F/30F difference will leak half as much energy as a 70F/30F difference. See where heating your house from 50 to 70 doubles your losses? Oct 5, 2021 at 2:04

3 Answers 3


If heat loss, and the associated amount of fuel required to add heat to the house to compensate for that loss, is the only concern then yes allowing the house temperature to track the outdoor conditions is most economical. Heat loss is proportional to the difference in temperature between indoors and outdoors. If the indoor temperature is held constant even while outdoor temperature drops, there's greater heat loss as compared to the case when indoor temperature tracks outdoor temperature with a consistent offset.

There are other complexities, like efficiency of running the heat system for intervals through the night vs continuous for an hour or more in the morning. For most systems a longer run is more efficient, but the difference diminishes quickly. A run of 10 minutes will be better for economy and for equipment life than a run of just 1 minute, but the difference between 10 minutes and 60 minutes will be minimal.

If the cost of your fuel varies through the day it's likely that will weigh far more heavily in the final bill than will length of cycles or nighttime set-back strategies.

  • The runtime efficiency has a lot to do with heating up the thermal mass of the system itself, and in many cases that heat eventually ends up in the living space anyway (unless your boiler/furnace and a lot of pipework/ducting is somewhere unheated). Short cycling isn't great for efficiency either, and this will happen if you're maintaining a temperature for a long period (modulating burners help there)
    – Chris H
    Oct 5, 2021 at 11:30

Is your question academic or specific to your home? For general & academics, there is a good discussion in a previous question linked in the comments by Isherwood: Why is turning down the heat during day and night considered more efficient?

If you are asking specific for your home, the best way is A/B split testing (as also @FreeMan suggests).

  1. Over a one month heating period split the days into two groups, A and B, e.g. odd and even dates. Generally you'd want them alternating, since whether changes gradually and you have a good chance of hitting the same or similar whether with two subsequent days and thus one of each split test. This way at-home holidays or weekends can be split between the two sets.
  2. On "A" days run the heat 24hrs, on "B" run your scheduled heating.
  3. Note your daily meter reading. Measure perhaps in the afternoon so as to capture night time usage plus any offset from morning solar heating. Also note the night time low and daytime high temperature from your weather forecast (outside thermometers are tricky to place due to sun, house heat and wind).
  4. Chart the consumption on A days over time throughout the month, chart the consumption on B days, chart the temperatures, and visually compare. If you wish you can update your question with the charts and we'll have a chance to comment on it.

There will be some variation due to outside temperature differences between an A day and the following B day, and your meter reading will be affected by other uses in the house too. But a 30 day experiment can on aggregate provide a reliable answer.

Also translate the difference in usage to actual financial difference. For this use the energy portion of your bill, not the total bill which often includes metering charges, delivery charges, connection charges etc... which are all fixed costs.

Now you have a dollar/euro comparison between the two approaches for your specific case.

Large savings also happen when you switch to a high efficiency furnace (from 50% to 95%), when you switch to the cheapest energy source in your region (gas vs electricity), and when you apply passive heating and insulation with air exchangers, solar capture etc... Then the question becomes which of those upgrades are actually worth the cost savings, but that's for a different post.

  • Would have not tought that it was a 'house to house' thing. Huh.
    – Fredy31
    Oct 4, 2021 at 19:22
  • @Fredy31, yeah I think your case could differ from someone across the country or across the street, even your personal schedule (kids/seniors, wake-up time compared to sunrise etc..) If you can test, you'd have a much better answer, and the gap could be theoretical but too small to really matter etc..etc.. Good question!
    – P2000
    Oct 4, 2021 at 20:03

In almost all cases it's more economical to allow your house to be closer to the outdoor temperature for as long as possible. At night or not at night, every degree and every hour of reduced differential reduces your bill. Re-establishing temperature later will not cost more, it will cost less because you lost less energy during the period of reduced differential.

Most heating and cooling systems are on or off. They run at one intensity only and their thermostats just turn them on and off to suit. If there exists some kind of heating, or air conditioning, that is able to run at a variable intensity where a lower output with a higher duty cycle is more efficient than running at high intensity and being switched on and off constantly, maybe there is something to investigate. I'm not aware of a system like that, at least not a residential one and I don't see why it would be more efficient even if it existed.

  • 1
    Modern condensing gas boilers (as used in the UK) can modulate down to less than 20% to maintain the circulating water temperature and exhaust temperature within limits. Mine is nominally 30kW but minimum 5kW. It appears to only run flat out for a few minutes before reducing the output. The reason that is more efficient is we want to keep the exhaust cool enough to condense; running hard briefly will very soon get the exhaust too hot. Stop-start operation and the resulting temperature cycling also increases wear
    – Chris H
    Oct 5, 2021 at 11:34
  • @ChrisH that's cool. It would be interesting to know how much more efficient it is to run that way. In a poorly insulated house on a cold day of course it would still make sense to let the temp drop, but maybe there is a balancing point.
    – jay613
    Oct 5, 2021 at 12:02
  • I was thinking that an 80% efficiency boiler will be 80% efficient regardless of what time it runs or on what duty cycle, and likewise a 90% one. But a 90% efficiency boiler that can run at 94% if water and flue temps are lower requires more thought.
    – jay613
    Oct 5, 2021 at 12:08
  • 1
    It's still going to be more efficient to let the house cool down. The gains from modulating are marginal in comparison, and more about keeping the 90%+ efficiency as demand reduces - if it stops condensing the efficiency drops fast. Mine has a thermostat and a recommended operating point that's just below the fall-off in efficiency. I run it a bit lower for a marginal improvement, turning it up in a cold spell mainly because some of my downstairs is a bit underheated. They're quite sophisticated these days
    – Chris H
    Oct 5, 2021 at 12:11
  • So what we need is N = L(b) / L(h). L(b) is energy leakage (not heating the house) in kW in your boiler at full output minus energy leakage at lower output. L(h) is energy loss of a house in kWh per heating degree day deduced from the weather forecast and experience with the losses for any particular house. If N > 1 you turn down the stat at night. (This is just for amusement, I'm sure it's never > 1)
    – jay613
    Oct 5, 2021 at 12:13

Not the answer you're looking for? Browse other questions tagged or ask your own question.