Become a Member

Get access to more than 30 brands, premium video, exclusive content, events, mapping, and more.

Already have an account? Sign In

Become a Member

Get access to more than 30 brands, premium video, exclusive content, events, mapping, and more.

Already have an account? Sign In



In Defense of the European Death Knot

Enjoy unlimited access to Climbing’s award-winning features, in-depth interviews, and expert training advice. Subscribe here.

European Death Knot Flat Overhand EDK
The offset overhand knot, also known as the European Death Knot.@OceangoingMonkey

I’ve read and listened to a lot of misinformation and misgivings regarding the European death knot (EDK) in climbing circles of late. Despite the grim nomenclature, the EDK is a safe and effective rappel knot and I’ll explain why, but first let’s take a short look at the whys and wherefores of the much maligned EDK.

What is it?

The European death knot is more correctly referred to as the offset overhand bend. Quite simply, it is an overhand knot tied using two lengths of cordage. It is used as a method to connect two ropes for use in rappelling.

What’s in a name?

Perhaps most of the uneasiness attached to the EDK comes from the name itself. Although the origin seems to have been lost to the mists of time, the name was likely bestowed by American climbers when introduced to the offset overhand bend by Europeans. It is also commonly referred to as a flat overhand bend, although more than one expert has stated that the term “flat” is nebulous and misleading, whereas “offset” has a specific meaning—that both standing ends converge initially and then follow a parallel path through the core of the knot.

Like many things in climbing, the EDK is counter-intuitive—it appears to be less safe than the other variants, when the inverse is actually true. Only the European part of the name stands up to any amount of scrutiny. The EDK is in fact a bend, the name given to methods of joining two ropes together, and is not likely to result in death.

Another problem associated with the EDK is that it is often conflated with the offset figure eight, an ostensibly similar bend which performs very differently. This has confused the reporting of failures.

EDK vs Offset Figure Eight

Often called a “variant EDK”, an offset figure eight is by no means a offset overhand bend. The method for tying them is very similar, except that instead to tying an overhand, one ties a figure eight.

The best data we have on the failure strengths of the EDK vs the offset figure eight come from a series of tests performed by Thomas Moyer in 1999. He also provides links to a few other organizations which have run similar tests.

In all fairness, Moyer’s opinion is that there are better knots than the EDK. This may be true if only strength is considered as a factor, but there are other considerations that I will elaborate on later. The upshot of his research is that a properly tied, dressed and pre-tensioned offset figure eight fails under significantly less load than a properly tied, dressed and pre-tensioned EDK (750 lbs vs 1400 lbs).

Arguments against the EDK

The method of failure for both bends is by the “capsizing” or “rolling” of the knot. The theory goes that, under load, the knot can roll on itself until eventually rolling right off the end of the tail, separating the two strands.

The story goes that there have been deaths associated with the use of the EDK, however the evidence is anything but empirical. Four accidents and three fatalities have been attributed to EDK failure: Zion in 2002, Grand Teton National Park in 1997, Big Cottonwood Canyon in 1995, and Seneca Rock in 1994.

However, as discussed earlier, this has been the result of incorrect labelling of the offset figure eight, which was the bend used during three of the four incidents. One major US climbing publication incorrectly attributed two offset figure eight accidents to the EDK this year alone. Clearly, something is rotten in the state of Denmark.

Moyer states that the EDK is not the method of choice for tying slings for use as anchors, which seems like a different conversation altogether. However, he also states that he has attended only one rescue involving a stuck rappel rope, as opposed to five which were the result of rappel knot failure.

Arguments for the EDK

There are a few reasons why the EDK is a sound choice. The main argument for the EDK is that it is low-profile and asymmetrical. What this means is that it is less likely to snag in cracks or over edges, and has a tendency to turn inward to present the flat side of the bend to the rock edge. This reduces the potential for stuck rappel ropes when compared to other alternatives.

Additionally, the EDK is easy to tie, easy to untie (under normal loads), and easy to inspect. This cannot be said with much conviction for the other variations. For use in multiple rappels, it offers a significant time saving factor.

Moyer downplays the importance of these factors, stating his record of rescues (again, one for stuck rope, five for knot failure). Perhaps it’s situational—the speed and efficiency of knot tying and avoidance of stuck ropes hardly matters on single pitch sport, but on an alpine face with an incoming storm, it could be very important indeed.

His data is the smoking gun in proving that the knot is safe to use for its intended purpose, which is to say for rappelling. It must be stressed that the EDK, like any other knot, must be well dressed and pre-tensioned. This significantly improves the failure rate of the knot. If tied correctly, the rope is more likely to break before the knot catastrophically fails.

Moyer suspects that the incidents in question were due to user error rather than physical failure. He goes on record to say that “millions of rappels have taken place on these knots without failures” and that he doesn’t believe that “the flat-overhand will ever fail under body weight if it is tied well.”

How to tie an EDK

With all this talk about incorrect tying, let’s talk about correct tying. 

Here are the important points:

  • Leave sufficient tail—30-50 cm is ideal
  • Dress and pretension the EDK
  • A backup can be tied in the form of a second overhand above the initial knot. This increases the knot profile, but Moyer states that it maintains asymmetry and makes the knot harder to capsize.
  • Don’t use the EDK for tying sling/webbing/cordelette anchors. Moyer suggests the Tape/Water Knot or the Double Fishermans.
  • CHECK YOUR KNOTS. Inspect your buddies. Complacency=bad juju.

For more information, see the offset overhand bend on Animated Knots.

What about ropes with different diameters?

There is not much data in comparison to the equal strands testing, however the upshot is that, once again, the cordage generally breaks before the knot fails via rolling. Therefore, it is feasible to connect two strands of unequal diameter rope, and the limiting factor will be the tensile strength of the smaller rope. That said, it must also be stressed that back-up knots had been used (in the form of a second overhand directly following the initial EDK) and had tightened during the testing. There is no data that pertains to the EDK without back-up knots, so it’s hard to know how it would perform without reverting to plain old guess work…

Other commentators have stated that it is a matter of importance to tie the smaller strand below the larger strand, which is to say that the smaller strand should present toward the rock when weighted. The theory goes that it is harder for the smaller strand to roll over the larger and thus initiate failure than the inverse. Again, this theory has not been definitively tested and no data exists, but it does appear rather intuitive and doesn’t seem to pose any disadvantages.

In the absence of better data the answer is—Yes, you can join ropes of varying diameters, but you should also employ a back-up overhand knot and ensure you tie the smaller diameter cordage under the larger.

Alternatives to the EDK

Even though it has proven itself time and time again, if the EDK still gives you the heebie jeebies, there are stronger alternatives. The cons of these are that they are harder to tie, untie, and inspect, although they do offer additional strength.

The double fishermans is the go-to alternative for many climbers, but can be notoriously hard to untie after being loaded and can easily be tied incorrectly.

The Flemish bend or figure eight bend (the follow through, not to be confused with the offset figure eight bend) can be used, however it MUST include stopper knots on each side. The absence of the stopper knots can allow the knot to untie itself under opposing loads. It remains relatively easy to untie but has a huge profile in comparison to the EDK.


It behooves us to check our knots. That seems like common wisdom, but it ain’t so common. The watchwords here are “well dressed” and “pre-tensioned.” The science is all there—the EDK is suitable for rappelling if performed correctly.

Should we rename the European Death Knot? Why bother, nothing else will ever stick. Death Knot sounds pretty fucking metal, and who am I to argue with that?