Thoughts on redundancy in climbing anchors
This article was written with editorial assistance from Richard Goldstone, thanks Richard!
Redundancy in anchors, while a good maxim for most climbers in most cases, is a situational and not an absolute rule.
One of many single piece, non redundant anchors on the Matterhorn. Photo by Dale Remsberg.
Single point, non redundant anchor on the matterhorn. photo: Dale Remsberg
“Good judgment comes from experience. Experience often comes from surviving bad judgment.”
Will Rogers
One of the core concepts of anchor building is redundancy - if any single component were to fail, the entire anchor wouldn’t fail. This mandate has been around for a long time for a good reason, because it's arguably the single best thing relatively newer climbers can do as a buffer against the mistakes of inexperience. The sketchier the situation, the more redundancy counts. When in doubt, double up isn’t a bad rule to live by. If you have the gear and the time, do it.
Sliding X anchor? Not redundant. If that sling gets cut, adios.
Statically equalized cordelette style anchor? Perfectly redundant, if any part of the sling fails, the rest of the anchor still works.
However, outside of anchor construction, climbing has many mission critical components that are not redundant.
Most of us climb on one rope
Pretty much every harness has one belay loop (Yes, it's doubled over and sewn, but it's still one piece of webbing)
We belay and rappel with one carabiner, with one belay device
We clip the one bolt/hanger with one carabiner/draw as the first clip on a sport route, and would deck if it fails
You get the idea. Why do we accept so many potential “single points of failure” in many parts of the system, yet demand it always be a component of our anchors?
The answer is, it doesn’t always have to be. Rather than being a mandate for every anchor, all the time, think of redundancy as a concept that applies in varying degrees to varying situations. It’s overall a good idea, and you should never be faulted for doing it, but it’s a situational, and not absolute, rule.
IFMGA Guide Dale Remsburg writes: “The idea of redundancy comes from pieces in the rock, not the links or tools we use to connect them.”
Keep in mind, creating redundancy comes at a cost of time, or gear, or both. Mostly, that’s a good tradeoff to make. Other times, perhaps not.
A fundamental principle of economics (and many other aspects of life, including anchors) is the law of diminishing returns, which, in econ-speak, means that adding additional factors of production eventually results in smaller increases in output. Say that it takes one builder one year to build a house. So, if you have 365 builders, can you build a house in one day? Of course not, because after a certain point, the extra production (builders) result in lower output (less work getting done because they are tripping over each other).
And yes, this can apply to building anchors. Continuing to add “production” (additional slings, backups, double locking carabiners, etc.) at some point do not significantly increase your “output” (safety margin), so it's probably not so smart to keep doing it. Of course, the question becomes, where do you cross that point? There's no firm answer, but here's one way to think about it mathematically.
Say that the odds of failure of an anchor sling are one in 1,000. If you backup that sling with another, the theoretical odds of anchor failure become 1,000 x 1,000, or one in 1 million. Add a third sling? The theoretical odds of failure are 1,000 x 1,000 x 1,000, or one in 1 billion.
What's your acceptable level of risk? If you're feeling good with one in 1,000, then going to one in 1 million (let alone 1 billion) is probably not worth doing.
Another broad component of anchors is having proper context in anchor photos/examples. This gets more into the situational judgment of when redundancy is more important. It's tricky to simply show a photo of an anchor and ask if it's acceptable or not. The slightly snarky yet truthful answer is, “It depends! Some broader context questions might be:
Is the anchor for a multi pitch lead belay, top rope anchor or a rappel anchor? Loads vary a LOT between these. Peak force on a rappel anchor, 2-3 kN; theoretical max force on a lead anchor, about 9kN. (Source)
How difficult is the pitch below or above, and what’s the skill level of the climbers?
Is it a part of multi pitch climb or the top of a climb?
Is there a chance of rock or ice fall that might hit the anchor?
Is it a casual day at your local crag, or are you trying to do a remote 15 pitch alpine route?
And so on.
Let’s examine that last point - that of speed/efficiency. When people cut corners on anchor building, the usual rationale is that it saves time and/or uses less equipment. Let's be honest though, the requirements for “speed” on a relaxed three pitch climb on a sunny day at your local crag (where time, weather, and daylight are not major issues) is minimal. On a casual route with no real need to hurry, eliminating redundancy from an anchor because it saves you a few minutes is not a very compelling argument.
How about a long, committing, 15 pitch alpine mountaineering route? There, shaving off a few seconds wherever you can might become more important, because the time savings multiply over the course of a big day.
Some people always seem to be in speed mode, striving to do everything the quickest way possible. If that works for you, great. But, for most of the rest of us, and especially for newer climbers, safety should always be a priority over speed.
Let's have a look at a few anchors that question the concept of redundancy. How do you feel about them? Consider the trade-offs in time, gear or both to make the anchor “textbook” redundant.
Rappel anchor
One tree, one sling, one carabiner. Every part of this anchor is non-redundant.
However, each of these three components is vastly stronger than any possible rappel load, which at most, even with terrible rappel technique, is never going to exceed 3 kN.
The tree is well rooted and stout enough. (Some folks use “5 & alive” for tree anchors - at least 5 inches in diameter, and alive.)
Some brand new 1 inch webbing (rated about 18 kN) tied in a well-dressed water knot with good long tails. The webbing is rubbing on tree bark, not the sharp edge of a rock.
Snapgate carabiner (rated 22 kN) left behind for the rope connection, gate taped shut for extra security (aka cheapskate locker). Extra points for using pirate hockey tape.
So, whaddya think? Would you rap on it? Why or why not? If not, what would you change / add so you’d feel comfortable?
How about this anchor? It’s a Fixe PLX/Duplex anchor, standard in many parts of Europe and Canada, but not so common in many parts of the USA. Everything is stainless steel, the rings are 10 mm thick, and the whole thing is rated at 30 kN.
How about that bottom ring? The gear testing wizards at HowNot2.com tested a couple of these. One broke at around 90 kN, the other broke around 60 kN! That is miles stronger than your rope, your belay device, your belay carabiner, and all those other single points listed above. If you're happy with those single points, why not be happy with this?
This is an anchor that can leave redundancy advocates scratching their noggin. Hmm, what do I do with this mishmash of hardware? Am I supposed to clip just that ONE ring at the bottom?! That’s not redundant, if it fails, YGD! (You’re Gonna Die).
Redundancy advocates might just ignore the chains and ring, and rig this with a long runner clipped to the bolt hangers. Remember, it's probably totally fine if you choose to do this, but it’s not the intended nor most efficient way to use this style of anchor.
Here’s an Instagram video posted by AMGA Certified Rock Guide Cody Bradford using this exact style of anchor. Cody clips a single large locking carabiner to the ring making a master point, then clips two carabiners onto that, one for his clove hitch and one to belay his partner. Yep, everything off the one ring. (And then everything off of one yellow carabiner.)
Not textbook redundant, right? What do you think about this anchor set up? What would you do with anchor hardware like this if you had to belay up your partner?
Here’s another example. I first saw this anchor on the Facebook feed of Dale Remsberg, an IFMGA Certified Guide and technical director of the American Mountain Guide Association. (So yeah, Dale knows his stuff.) It's a photo he took of an anchor he built while guiding a client, and put it on Facebook to start a discussion about, guess what, redundancy. (For context, it’s on a large ledge, the pitch below is an easy 5.7ish, and there is no risk of rock fall from above.)
image: Dale Remsberg - https://www.instagram.com/p/BxH139Tj1M5/
What do you think of this?
At first glance, redundancy advocates would dismiss this immediately. A basket hitched sling has zero redundancy; if it gets cut or fails, immediate anchor failure and YGD!
Technically true, but how would this sling possibly fail? It's running around tree bark, not any sort of sharp rock. There is zero risk of rockfall from above impacting the sling. The only way it could fail for is for it to physically break, something that has pretty much never happened in the history of rock climbing outside of the ever awesome Sly Stallone movie, Cliffhanger. (Readers, please correct me on this last sentence if I’m mistaken.)
One approach to make this anchor redundant is to tie a knot in the sling. Maybe so, but what's the trade-off? You might barely have enough sling material to make the knot . . . or perhaps you wouldn't. You also weaken the sling by tying a knot. You also take the time to tie the knot, and probably a longer time to untie it, which if it gets weighted, could be a hassle. (Side note: could be a great place for a girth hitch at the master point.)
Which is more important, having the full strength of the runner, or weakening it by tying a knot which creates redundancy? Oh, and there’s just one tree branch there, that's certainly not redundant. What about that?
Now, I saved that's the best for last. Here's one that’ll give redundancy advocates nightmares. (And to be honest, I’ve never seen one of these in real life and I'm not super excited about it either . . . )
How about ONE SINGLE BOLT?!
Here’s a screen grab from a video, link below. From the video: “( . . the anchor can) . . . in some instances be a single large glue in bolt, which is the only anchor at the anchor point.”
How are you going to make this redundant? Answer is, you probably can't.
image : screen grab from: https://youtu.be/1r7hIZREJoQ?t=
Now, before you start thinking this is a 20 year old photo from East Boondockistan, this video comes courtesy of the excellent (if awkwardly titled) “Safety Academy Lab Rock” video tutorial series, produced by the well-regarded German company Ortovox, and backed by Petzl and the German Mountain and Ski Guides Association (in German, “VDBS”). So, while if Americanos like me may not have any personal experience with it, the fact that it’s featured in an instructional video made by VDBS I’d say give it a fair bit of cred.
And no, you don't truly know the quality of the steel, the length of the bolt, the type of adhesive used to glue it into the rock, etc. But, a properly placed long glue in bolt like this has a UIAA minimum standard downward pull of 25kN, and have actually tested up to 50(!) kN, which makes it about the strongest component you'll pretty much ever encounter in climbing (right up there with the huge master point ring in the Fixe anchor above.) So, the short answer is yes, you can probably rely on this single point of connection. (But, in all honesty, as an American climber raised with the mantra of redundancy, I would not be overjoyed to discover this as my only connection to the rock.)
But hey, if you find this at the top of the first pitch and you don't like it, you can always rappel off and go climb somewhere else, right? :-)
Below is the whole video; see the single point anchor part starting at 0:40.
(And the fixed point belay? We’ll cover that soon in another Alpinesavvy article.)
And, for a little historical perspective, here’s one of many single piece anchors on the iconic Matterhorn, photo by Dale Remsberg.
Stout? Looks like it.
Used by probably tens of thousands of climbers over decades, most of them professional guides? Yes.
Redundant? Nope.
anchor on the matterhorn. photo: Dale Remsberg
