When I bought my first real sleeping bag, I looked at the tag, saw a number, and treated it like a promise: “rated to 20°F” meant “good to 20°F,” end of story. Then I had the cold night that started this whole site, went back and actually learned how those numbers are set, and realised I’d been reading the wrong one. If you take one thing from this post, take this: most bags list two temperature numbers, and the warmer-sounding one is not the number you should plan around.
The two numbers on the tag
Modern sleeping bags are tested to a standard called ISO 23537 (it replaced the old EN 13537 standard, and you’ll still see both names floating around). The lab puts a heated mannequin in the bag, dressed in a base layer, on a standardised pad, and measures heat loss. From that, you get a few ratings. The two that matter for buying are:
- Comfort rating. The temperature at which a “cold sleeper” — the standard models this as an average woman — can sleep comfortably, in a relaxed position, through the night. This is the realistic “you’ll actually sleep” number.
- Limit rating. The temperature at which a “warm sleeper” — modelled as an average man — can sleep for eight hours, but curled up, conserving heat, not relaxed. This is closer to “you’ll get through the night” than “you’ll sleep well.”
There’s also an Extreme rating, but ignore it entirely for planning. It’s a survival figure — the temperature at which a sleeper is protected from hypothermia for a few hours. It is not a “rated to” number in any useful sense, and no honest setup should ever lean on it.
Why the Limit number is the one that burns people
Here’s the trap. Marketing loves the lower-sounding, more impressive number. A bag whose Comfort rating is 30°F might have a Limit rating around 20°F, and guess which one gets printed big on the stuff sack? The 20°F. It sounds warmer, tougher, more capable. So a beginner sees “20°F bag,” checks that the forecast low is 25°F, feels safe, and goes to bed cold — because at 25°F that bag is only just into its Limit range, where you survive curled up, not its Comfort range, where you actually rest.
That’s almost exactly what happened to me. The optimism in that big printed number, combined with a cheap pad underneath, is a reliable recipe for a sleepless night. (The pad half of that story is its own post — the night I froze — because warmth is always the bag and the pad together.)
How I read a bag tag now
My rule is simple and a little conservative, and I sleep warm because of it:
- Find the Comfort rating, not the Limit. If a listing only gives one number, assume it’s the Limit (the flattering one) and mentally add 10–15°F to guess the Comfort figure, or treat the bag as rated 10–15°F warmer than it claims.
- Match the Comfort rating to your forecast low — with margin. I want the Comfort rating at or below the coldest temperature I expect, and then a little extra on top for a clear night, exposure, altitude, or just being tired and run-down at the end of a long day.
- If you sleep cold, add more margin. The Comfort rating is built around a cold sleeper already, but everyone’s different. If you know you run cold, buy warmer than the math says.
I built the sleep-system matcher to do this for you — punch in your overnight low and your sleeper type and it returns a target Comfort rating with the margin already baked in, plus the pad R-value to go with it. It also shows you the rough Limit equivalent so you can see how much warmer the marketing number sounds than the reality.
”But buying to Comfort means a heavier, pricier bag”
Sometimes, yes. A warmer bag usually costs more and weighs a bit more, and as a gram-counter I feel that. But this is exactly where warmth-to-weight thinking earns its keep. A high-quality down bag bought to the right Comfort rating can be lighter than a cheap bag you’d have to over-buy or layer up inside. The goal isn’t to carry the warmest possible bag — it’s to carry the right warmth for the least weight. If you’re weighing two candidates, the warmth-to-weight comparator will tell you which one gives you more warmth per ounce.
What the rating doesn’t tell you
One more thing worth saying plainly: even the Comfort rating is a lab number, and the lab is not the trail. The ISO test runs on a standardised pad, with a base layer, on a heated mannequin, in still air. Your night has none of those controls. A clear, windless sky radiates your heat away faster; altitude makes the air colder than the valley forecast; a damp bag lofts less; an exhausted, under-fed body generates less heat; and a draughty quilt or a cold pad undercuts the whole thing. None of that is in the rating.
So I treat the Comfort number as a starting point I adjust from, not a guarantee I lean on. That’s the entire reason I buy with margin and pair the bag with the right pad rather than trusting the tag — the rating gets me close, and the margin covers the gap between the lab and the actual night. Anyone who tells you a bag is simply “good to X degrees, full stop” is reading a controlled test as a promise it was never meant to make.
The bottom line
Two numbers on the tag, and the marketing pushes the wrong one. Buy to the Comfort rating, match it to your forecast low with margin, add more if you sleep cold, and never, ever plan a trip around the Limit or Extreme figure. Do that and the bag stops being a gamble.
When you’re ready to actually pick one, I keep a spec-led roundup of the best backpacking sleeping bags, sorted so you can see the Comfort ratings up front instead of hunting for them.
These are general observations from my own trips, not lab measurements or a prescription for your specific gear. Temperature ratings are lab figures (ISO 23537); real-world warmth varies with the sleeper, the pad, the shelter and the conditions, and no rating is a survival guarantee.