Sleeping bag temperature ratings, decoded

A sleeping bag rating is not a promise about you. Under ISO 23537 it is the output of a heated copper manikin, wearing a specified base layer, lying on a specified mattress, inside a bag with a specified amount of air in it. The standard produces three temperatures from one test, and the one printed largest on the hangtag is usually not the one you should buy on.

What ISO 23537 actually measures#

The test is a steady-state heat transfer measurement, not a night's sleep. A copper manikin is heated to a fixed skin temperature and instrumented in segments. It is dressed in one long-sleeved base layer and a pair of socks, laid on a mattress with a defined thermal resistance, and put in the bag inside a climate chamber. The lab measures how much electrical power it takes to hold the manikin at temperature as the chamber gets colder.

From that single power curve, the standard runs a physiological model to produce three numbers:

The gap between comfort and limit is not a fudge factor, it is two different physiologies and two different postures modelled from the same measured insulation. Women, on average, have a lower resting metabolic heat output per unit of skin area, and the model reflects that. This is why the same bag can honestly be described as a "0 degree bag" or a "minus 6 bag" depending on which line you read.

Why the rating is a system rating, not a bag rating#

The detail most buyers miss: ISO 23537 specifies the mattress under the manikin. The bag is not tested in isolation, because it cannot be. Roughly a quarter to a third of your heat loss overnight goes downward into the ground, through insulation that your body weight has crushed to nothing.

That means the number on the bag assumes a pad of a given thermal resistance. Put the same bag on a thin summer air mat and the rating is no longer describing your night. Our sleeping pad R-value guide works through the conduction maths, but the short version is that heat flow to the ground scales with 1 over R: going from R-1 to R-2 saves you about 25 watts, and no bag upgrade compensates for a pad that is one full season short.

The same logic runs through every other variable: your dinner, your hydration, the wind on the flysheet, whether you are on night one or night five of a humid trip. That is what our sleep system calculator exists to add up.

Converting any number into your number#

Here is the conversion we use, and the one the calculator applies. Start from the label, get to a comfort-equivalent figure, then adjust for the system and the sleeper.

1. Find your starting point. If the bag quotes an ISO or EN comfort rating, start there. If it quotes only a limit rating, add 6 degrees C. If it quotes no standard at all, add 8 degrees C and treat the result as a rough estimate.
2. Check the pad. Compare your pad R-value with what the temperature demands: roughly R-2 at 10 degrees C, R-3.4 at 0, R-5 at minus 10, R-6.5 at minus 20. Every full R-value you are short is worth roughly 2 degrees C of felt cold.
3. Adjust for the sleeper. People who consistently sleep cold should add about 3 degrees, which is the same size as the gap the standard itself reports between its two manikins.
4. Adjust for the night. Sleep clothing, a liner, wind exposure, a hot meal, accumulated damp in the fill.
5. Demand a margin. Aim to end up 3 degrees C below the forecast low, not level with it.

The margin nobody budgets for: moisture#

A rating is measured on a dry bag. Over consecutive nights, insulation picks up moisture from your body (a sleeping adult gives off roughly 40 to 70 grams of water per hour) and from the air inside a shelter. Down loses loft as it absorbs moisture, and loft is warmth.

On a humid multi-day trip without a chance to dry the bag, expect the effective comfort temperature to drift 3 to 5 degrees warmer by night three, which is to say you will feel colder. The countermeasures are cheap and boring: air the bag at every lunch stop that offers sun or breeze, use a vapour barrier liner in deep cold, and keep the bag out of contact with tent walls. Synthetic fill degrades less in this respect, which is a large part of the case in our down against synthetic comparison.

Test it yourself: the two-thermometer method#

You do not need a lab to sanity check a bag. You need two cheap digital thermometers, one on a probe.

1. Pitch or lay out the bag on the pad you actually use, in the shelter you actually use, on a night at least 5 degrees below the bag comfort rating.
2. Put one thermometer outside the shelter, shaded and off the ground, to log the real overnight low. Campsite lows differ from forecast lows by several degrees.
3. Run the probe of the second thermometer to the inside of the bag near your chest, taped so it does not sit against skin.
4. Sleep. In the morning, note the outside low, the inside temperature, and the honest answer to one question: did you wake up cold, and at what hour?

Three nights of this tells you your personal offset from the label, which is the most useful number in your gear notebook. It also tells you whether the problem is the bag at all: a bag interior sitting 25 degrees above ambient with a cold back is a pad problem, not a bag problem, and no amount of new bag fixes it.

What a rating does not cover#

• Fit. A bag that is too long for you has extra air the body must heat, worth around 2 degrees. A bag too narrow compresses insulation at the shoulders and hips.
• Draught features. Draft collars, zip baffles and hood cinches are where the last few degrees live. Two bags with identical fill weight can differ by several degrees on these alone.
• Movement. The manikin does not roll over. If you sleep on your side and turn six times a night, a mummy bag that rotates with you behaves differently from one you rotate inside.
• Recovery. The test is one steady state. It says nothing about how the bag performs on night four after two damp nights.
• Ageing. Fill loses loft with use, compression and body oils. A five-year-old bag that has never been washed is not the bag that was tested. Our down washing guide covers restoring it.

Common mistakes we see repeatedly#

Buying the extreme number. It appears on some retail listings as if it were a usable rating. It is a survival limit involving shivering, and using it as a purchase basis is how people end up in genuine trouble.

Comparing an ISO bag with an unrated one on price. Testing costs money, so cheap bags often skip it. The unrated bag is not necessarily bad, but its number was chosen by someone with an incentive, and it should be treated as a marketing figure until proven otherwise.

Ignoring the pad. By far the most common cause of a cold night in a correctly chosen bag. See the R-value guide.

Treating a quilt rating as a bag rating. Quilts have no back and no hood, and their ratings are almost never ISO tested. See quilts against sleeping bags for how to compare the two honestly.

Assuming women's bags are a marketing exercise. They are usually cut with more insulation at the feet and hips, and rated to comfort rather than limit. The specification difference is real, whatever you think of the colours.

Frequently asked questions#

Is the comfort or the limit rating the right one to use?#

Comfort, for almost everybody. It is the temperature at which the standard sleeps rather than endures, and it maps to how most people expect a bag to behave. Use the limit rating only if you know from experience that you sleep warm, and even then keep 3 degrees of margin over the forecast low.

How much colder is a bag on night three of a wet trip?#

Expect 3 to 5 degrees C of lost performance in down fill after two or three humid nights without drying, and less in synthetic. The loss is gradual, so people rarely notice it happening and usually blame the weather. Airing the bag for an hour at midday recovers most of it.

Do sleeping bag ratings work below minus 20 C?#

The standard covers the test method rather than a temperature ceiling, but the physiological model behind the comfort and limit outputs is least reliable at the extremes. Below about minus 20 C, expedition practice is to work from fill weight, loft measurement and layered systems rather than a single rating, and to treat any published number as a starting point.

Why do two bags with the same fill weight have different ratings?#

Fill power, baffle construction, cut and draught features. The same 500 g of down at 850 fill power lofts about 30 percent higher than at 650, and a sewn-through baffle pinches insulation to zero along every seam. Fill weight tells you how much insulation is present, not how much of it is working.

Are men's and women's ratings on the same bag comparable?#

They come from the same manikin run: the comfort figure models the standard woman, the limit figure models the standard man. So a bag with comfort 0 and limit minus 6 is describing two people in the same bag, not two different bags. Compare like for like, always.

Does a bag liner really add warmth?#

A silk or thin synthetic liner adds around 1 degree C, a fleece or thermal liner around 3. Both keep body oils out of the fill, which protects loft over years. If a claim on a liner package says 8 or 10 degrees, treat it as marketing: no 150 gram fabric sleeve does that.

What temperature rating do I need for summer camping?#

For summer lowland camping with overnight lows of 10 to 15 degrees C, a bag with a comfort rating of about 7 to 10 degrees is right, paired with a pad of R-2 or better. Below that you are carrying insulation you will unzip and vent all night, and a bag that vents well matters more than one that is warm.