This is a topic often discussed online, but all I could find so far was people sharing their opinions (including numbers) on what tie-in knot is the best. Here's what I've found out from different sources:

  • Figure eight: Fool proof to check, very hard to open after a fall
  • Yosemite bowline: Not as easily checked as figure eight, strongest knot for dynamic rope tension, easy to untie even after taking a punch.
  • Bowline on a bight: Easy to check, almost as strong as yosemite bowline, very easy to untie (in my experience it never tightens at all in the first place, being loose unless actively pulled on).

While I'm not at all against personal opinions as answers, these are usually not very good advice just on their own when it comes to matters critical to security; it's your life hanging on that knot after all. I would especially appreciate data on the knot strength; all the values I've got so far come from this German Wikipedia article, where somebody who doesn't know how to quote wrote down some numbers.

Long story short: Which one to use? Figure eight is starting to annoy me when my hands are tired and I've fallen. :-)

  • Are you interested in the strength of each knot, or the usefulness? There are very, very few situations aside from massive shock loads that would be an issue for any of the three, strength-wise.
    – Greg.Ley
    Commented Feb 19, 2012 at 6:36
  • 2
    @Greg.Ley The problem is more that I want to get rid of the 8. I like the yosemite bowline because it feels tight and is said to be safe, but I've learned it from the internet and am a bit scared using it because of that. As for what I'm interested in: anything you can tell me about these. Possible ways of doing them wrong, what's the worst case that can happen and so on. And, as said, hard data on strength would also be nice.
    – David
    Commented Feb 19, 2012 at 6:50
  • 2
    Relevant information. The bowline is apparently more prone to being tied or used incorrectly.
    – Mr.Wizard
    Commented May 17, 2012 at 8:31
  • 5
    Your climbing partner needs to be able to check your knot. They're expecting a rewoven figure-eight, and they can reliably identify whether a knot is or is not a figure-eight. For those reasons, it's not a good idea to use some other knot for your tie-in, even if you understand it and think it has some slight advantage.
    – user2169
    Commented Apr 4, 2014 at 16:58
  • 1
    The best knot is the one you can reliably tie whilst being pelted by horizontal spindrift and balancing on one foot. Commented Jul 11, 2017 at 8:30

7 Answers 7


Most climbers use a re-threaded figure 8. The knot is not that important, though. In reading many accident reports, I have never seen one where the knot came untied or where there was a rope failure due to the knot on the harness. Pick a knot you're very familiar with, check it, and you're done. Spend more time checking belay devices, locking caribiners, communication with your partner, and especially your protection.

One way to untie a tight double figure 8 is to bend the entire knot back and forth, then push both parts (parallel strands) of one end into the knot. This should loosen it enough to come undone easily.

  • 6
    Double figure eight every time - it never fails, and is always easy to untie, even after a good drop.
    – Rory Alsop
    Commented Feb 19, 2012 at 12:47
  • 4
    @Rory the OP is already using a retraced 8 and finds it hard to untie. Are you suggesting a different kind of figure eight?
    – Mr.Wizard
    Commented Feb 20, 2012 at 1:05
  • 1
    @studiohack I think I made a valid comment: the OP is using the 8 and finding it hard to untie; stating "easy to untie" and leaving it at that isn't exactly helpful.
    – Mr.Wizard
    Commented Feb 20, 2012 at 1:43
  • 2
    @Rory please, is this a single or double 8? If that is a single, how do you tie a double?
    – Mr.Wizard
    Commented Feb 20, 2012 at 9:37
  • 1
    To tie in you tie a single figure of 8 in the rope leaving a long tail. Put the end through your harness and then create a loop and put the end back through the figure of 8 following the same path as the first. This video probably explains it better than me youtube.com/watch?v=G1GzrRYcrM8.
    – Phil
    Commented Feb 20, 2012 at 11:12

The three knots you listed are all quite suitable for tying in to a sit harness for rock climbing.

The figure eight is the most widely used and most readily and easily inspected, and is not a bad choice for tying in:

...When tying into the rope there’s a reason the figure eight knot has been the knot of choice for years. Strong, simple, easy to untie and nearly as strong as the rope, there is no better knot for connecting the climber to the system...

Tie the knot that you and your partner or group are absolutely confident in and can inspect without having to even think about it that hard.

You might think it's silly, but if you can't tie the knot with your eyes closed, you probably shouldn't use it until you can. Whatever the knot, it should become second-nature, and then you and your partner still need to check it.

  • “You might think it's silly, but if you can't tie the knot with your eyes closed, you probably shouldn't use it until you can.” Doesn’t really matter if it’s easy to check and you do check it. Which is the greatest advantage of the figure eight.
    – Michael
    Commented May 13, 2019 at 11:19

I would say, dismiss the bowline on a bight since it needs a backup stopper knot to prevent loosening over time so it's not a "complete" knot like the other candidates.

The bends on the yosemite bowline are generally less tight than the figure eight. This bowline is less likely to jam, wears the the rope involved in the knot less, so from a rope longevity, fall damage to the rope, and general stress sense, the yosemite bowline would have to be the superior knot from an "engineering" standpoint.

I would not choose a knot on that basis and instead choose the one my party is most comfortable tying over and over and checking over and over. I don't see how the technical superiority of the yosemite bowline would make it measurably better "in the field" than the figure eight. If I didn't have a fid to assist in knot untying or my finger strength / rope choice made "untieability" after falls an issue, that would be a fair reason to choose a bowline over a more simple and jam-prone knot.

  • 1
    The only arguments I have ever heard for the yosemite Bowline are from an "engineering" stance. But the ease of (tying and checking) the 8 make it the choice of most climbers. I agree with you. Commented May 23, 2012 at 22:40
  • 2
    The reason I switched to the (Yosemite) Bowline, years ago, was the figure eight's "untieability" after falls. It's easy to tie and to check as well. Commented Aug 22, 2012 at 7:30
  • 7
    Does a bowline on a bight really require a stopper? It's a stretch but imagine the working end slipped out of bight that holds it then kept slipping until the nipping turn was gone too. You would be left with a normal bowline with a really long tail - still very safe.
    – Jake
    Commented Mar 28, 2013 at 3:36
  • 3
    The bowline on a bight does not need a backup stopper. It is a complete knot, but it's good practice to back it up, as well as the figure eight. The standard bowline, however, is a different story. I wouldn't climb on it.
    – montane
    Commented Dec 15, 2014 at 2:46
  • @manoftheson the bowline is used for all sails I ever sailed with, and never failed. What are the reasons why you wouldn't climb on it?
    – njzk2
    Commented Nov 17, 2015 at 19:16

I have been climbing on a yosemite bowline for a long time, and never had any problems. I don't see it as any harder to check than a figure 8 once you are used to it. It's easy to tie, secure, faster than the figure 8, requires less rope, and never gets difficult to untie after falls. It's a great knot, and there seems to be a lot of superstitious aversion to it out there. I have been accosted by teenage gym workers on a couple of occasions who have almost no climbing experience about how "dangerous" the bowline is to climb on. At best it's nonsense, at worst it's insulting. Nothing wrong with the figure 8, but the yosemite bowline is a fine knot for climbing.

  • 3
    No objection to the bowline in its various forms, but word of caution on the Yosemite finish – dressed incorrectly, it will invert and become unsafe. This is a pretty clear demonstration: youtube.com/watch?v=1dj5Y3h1AEI
    – Matt
    Commented Aug 28, 2018 at 20:56

The F8 is an inherently secure eye knot. Not all knots are inherently secure - but the F8 is, and hence one of the reasons for its enduring popularity for use in life critical applications. The simple Bowline (#1010) has been known and used for hundreds of years by sailors - and it was 'invented' for this purpose (ie in sailing applications). The simple (#1010) Bowline is not inherently secure - but thats no alarming revelation - it was never intended for life critical applications!! So climbers realized that #1010 was jam resistant - in fact it has no known jamming threshold - and so they adopted it for climbing applications. Back in the day - they used to use vegetable fibre ropes which were quite 'frictive'. Modern synthetic ropes are quite slick by comparison. The Yosemite Bowline was an attempt to address the insecurity of the simple (#1010) Bowline. However, its 'security' is very dependent on the type of climbing rope used. The Yosemite Bowline does not perform so well in stiffer ropes (ie not so secure).

With regard to the #1080 Bowline-on-a-bight - this also suffers from certain vulnerabilities. These issues can be addressed by adding a 'backup stopper knot' - which is technically a strangle. However, and here is the key message to all... you are far better off using a tie-in knot that is inherently secure. As stated, the F8 (#1047) is inherently secure. A knot which has the property of 'inherently secure' is one that does not require a backup stopper knot (ie strangle) to lock it down. As dynamic climbing ropes get thinner and thinner, the propensity for jamming becomes an issue. Enter stage right - 'Bowlines'. Note: The F8 eye knot is vulnerable to jamming (at loads from 4kN). Bowlines are also Post Eye Tiable (PET) - the F8 isn't 'PET'. If you are going to tie-in with a 'Bowline' - you should select and use a 'Bowline' that is inherently secure (one that doesn't require any backup stopper knot).

Examples of inherently secure 'Bowlines' include: 1. EBSB Bowline; and 2. Harry Butlers Yosemite Bowline; and 3. Scott's locked Bowline; and 4. Lees link Bowline.

Each of the above 4 'Bowlines' are inherently secure and stable - and suitable for life critical applications. Obviously, you need to learn how to tie them correctly. Just like you learnt to tie the F8 (#1047) correctly! Training and practice is key to mastering all skills.

I will also comment on the notional concept of a knots MBS yield (ie strength - the load at which point the knot begins to yield). It is a myth - an urban legend. The MBS yield of a knot is entirely irrelevant in climbing. There is no load that a falling climber can generate that will reach the MBS yield point of a knot. So why is strength always appearing in knot discussions? Because information is spread by word-of-mouth and few actually stop to question it. When people talk of 'strength' - what they really mean (but don't understand fully) is security and stability. What matters most with knots used in life critical applications is security and stability (not MBS yield).

NOTE: The image of the tying sequence shown above is not a #1013 Double Bowline. It is in fact a #1080 Bowline on-a-bight. Its yet another example of misinformation / poor journalism. Furthermore, both the #1013 and #1080 Bowlines are not inherently secure.

  • Another advantage of the figure eight over the bowline is that it is easy to verify that it is tired correctly. There have been several cases where an incorrectly tired bowline came undone and the result was serious injury or death.
    – Qudit
    Commented May 10, 2019 at 20:40

From my personal experience, people either use a double bowline (*) or a double figure 8. The bowline (in my experience!) is used only by fairly experienced climbers, probably because it is more difficult to execute.

I will list some pros/cons of the two knots; remind anyway that both knots are absolutely safe to use (as I'm sure is the Yosemite bowline, which I don't know).

Double Figure 8


  • Easy execution
  • Easy to check


  • The rope tightens up after a fall, making loosening difficult. For the same reason, it will ruin the rope faster than the bowline.

Double Bowline


  • Doesn't tighten up after a fall (so it also ruins the rope less)


  • It has a tendency to loosen up, so a security knot is needed to prevent this (figure F below).
  • More difficult to execute than the figure 8, so less beginner-friendly.
  • Trickier than figure 8 to check.

enter image description here

Execution of a double bowline with security knot. Figure taken from Alpinismo su Roccia - Club Alpino Italiano (2008)

(*) I think it is equivalent to what you refer to as "bowline on a bight".

  • 1
    The bowline is also more difficult to check that you've done it correctly, where as it's obvious when you tie a figure eight wrong. That's the main reason for the shift away from the bowline. I know you put this in the pros for the figure eight, but it's an important con for the bowline too
    – user2766
    Commented Jan 15, 2018 at 13:28
  • 1
    @Liam Mmh, you are right. Don't use the bowline so much myself, so I thought that it was fairly easy to check, but I guess that it still is more difficult than checking a figure 8. I edited the answer to clarify.
    – valerio
    Commented Jan 15, 2018 at 13:36

As far as being able to tie a knot blindfolded, etc... A better measure is to understand how knots work, and what constitutes a properly tied knot. If I was half way up a big wall and you insisted that I tie in using no known knot, and a substitute that was minimal and elegant, I would have no difficulty coming up with one. You can do it by rote, or you can actually understand what is going on.

It is not true that any of these knots is nearly as strong as the rope they are tied in. They run 65-80% last I heard, which was a long time ago. However, dynamic ropes never develop impact loads that threaten overall tensile strength (if the parts are healthy) so it works out. I figure the stronger an otherwise functional knot, the longer the rope end will last since the loads fall in much the same place every time. Can't hurt anything to use strong knots.

I think there may be sufficient experience in the guide and gym community to throw legitimate shade on the bowline, when dealing with essentially non-climbers. But there is also sufficient evidence to suggest that having two belayers on the same rope is a rational process (in series, not at either end as in normal roped climbing). So what needs the insurance company, and what needs the real world are two different things. Your laces are pretty important when climbing, and one has to wonder what insurance company tie is acceptable there. Belt or suspenders, boxers or briefs. All stuff we need the insurance company to weigh in on, and in certain circles, they would have legitimate info.


The thing about the figure of 8 being easier to check and all that is fine if you are not in any way interested in anything other than the simplest climbing set-ups, which are adequate for a great deal of climbing. But in anything more complicated, you better be able to do more than follow nursery rhyme sequences.

People I meet tend to divide into people who are technical and/or good knot tiers, and people who are useless, and the division is far from equal. So really there are two sets of instructions. One for those who are hopeless, and one for those who have strong skills.

Though it isn't the metric I use. Honold uses the 8 with follow through in the videos I have seen. I am a double bowline guy.

  • 3
    Even though I repeat myself: I don't see how this answers the question or adds significant value to the existing answer (part of it should be a comment on another answer, which you can do once you gained some reputation). Stackexchange is not a forum to exchange views, it's strictly for clear answers to concrete questions. Also many answers to this particular question are not a perfect example of the standards we'd like to have, containing mostly opinions/anecdotal evidence, so maybe don't take them as an example.
    – imsodin
    Commented Jul 11, 2017 at 7:37
  • Your penultimate paragraph is a key observation, and it applies to most aspects of technology, including knots - some people are rigorous rote learners and some are intuitive improvisers. If you want to explore this idea further, consider reading Zen and the Art of Motorcycle Maintenance by Robert M. Pirsig, who has much to say on the subject. Commented Jul 11, 2017 at 14:46
  • This post has been flagged as "low quality". As I agree that it is not a prime example of an answer, I don't see how this is "low quality". The content is well explained, well structured and the length is sufficient. I decline the flag.
    – OddDeer
    Commented Jul 12, 2017 at 8:18
  • People I meet tend to divide into people who are technical and/or good knot tiers, and people who are useless the last words of someone who didn't check their knot correctly.
    – user2766
    Commented Jan 15, 2018 at 13:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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