Common sense can get you killed
In front of me I have two books from Falcon's How to Rock Climb series: Self-Rescue by David J. Fasulo (published 1996), and Climbing Anchors, 2nd edition, by John Long and Bob Gaines (published in 2006 - the first edition, by Long alone, was published in 1993).
Fasulo starts out by saying that all the techniques described in his book assume that your belay anchors are completely solid. Based on the first edition of Long's book, he recommends a rigging method called a "cordelette": a bigass loop of cord which you clip to all the gear in your anchor, then draw together in a big overhand knot and clip into.
Brutally simple¹, extensionless, equalised by construction, and only requiring a bigass sling, which is a good thing to have anyway: what's not to like?
He also describes a system called a "sliding X":
but almost all the illustrations in his book show cordelettes.
He also discusses the best material for making cordelettes: "5.5mm Blue Water Titan Cord is tested at 2200kg, while standard 7mm accessory cord tests at 900kg. Obviously, the 5.5mm Titan cord is better suited, though because of its slippery nature [it] should be tied with a double or triple fisherman's knot."
Obviously. It's common sense, right? 2200kg is clearly better than 900kg in this case, and the fact that the 5.5mm cord is lighter makes the choice even clearer.
Unfortunately, this is one of those cases where common sense can get you killed.
For the second edition of their book, Long and Gaines had an extensive series of lab tests done, and found that in terms of equalisation the cordelette is actually the poorest performing system tested, particularly (as in almost all real-world climbing situations) if the "arms" of the cordelette are of unequal length. The cordelette's performance becomes particularly awful (with a difference between arms averaging 4.5kN) when it's rigged with high-test cord.
But why could this be? Well, watch this suggestive (and terrifying) video of some tests performed by the Welsh gear firm DMM. Their commentary on the video is here.
tl;dw: Dyneema (basically the same stuff as the high-test cord mentioned earlier, and with similar equalisation performance) has very high static strength (the amount of weight it can hold up at rest), but much less give and stretch than ordinary nylon. This means, surprisingly enough, that it has lower dynamic strength: it's easier to break it by dropping a heavy weight (such as, say, a climber) onto it. Moreover, it transmits more force onto its attachment points. This lack of elasticity probably explains the poor equalisation, too. There's a more detailed investigation of various high-strength cords at Tom Moyer's site, which is clearly a site I'm going to have to devote some time to reading.
Long and Gaines now recommend a hybrid system called the "equalette", which combines most of the simplicity of the cordelette with the self-equalizing properties of the sliding X: the tests show that this outperforms both the older systems. For my part, I've always tied in to the anchor directly with the rope, as the DMM guy advises; this is more fiddly than tying in with a cordelette, and suffers from having a preferred direction of loading, but introduces more dynamicity into the system. Unfortunately, it's not one of the configurations that Long and Gaines had tested; looks like it's time for me to hit up Google Scholar.
How many people died because of this bad advice (which was and is widespread: I've unfairly singled Fasulo out, but it's repeated in many other books). Probably not many, and certainly fewer than were saved by the good advice in the books. As Long and Gaines repeatedly point out, anchor failure's a rare event, and anchor failure due to choice of rigging strategy is even rarer: it's much more important to ensure that your actual gear placements are solid. But probably one or two.
The real question is "why was this advice promulgated before the lab tests were done?" I'm guessing the answer is somewhere between "worl, s'obvious, innit? Why bother testing?" and "lack of money". As the sport expands, more money for testing should become available; and hopefully this kind of thing will convince the climbing community that while common sense may be enough in everyday life, it can be a woefully inadequate guide when lives are on the line.
This is, I think, a general principle. Reliance on common sense can get you killed. If lives are on the line, do science whenever possible.
¹ As legendary Himalayan climber Ed Viesturs points out in this fascinating interview, tiny errors can have fatal consequences in the mountains. Simple is good in mountaineering.
Note: as should be obvious, climbing and mountaineering are activities with a danger of personal injury or death, and participants in these activities should be aware of and accept these risks and be responsible for their own actions and involvement.
Fasulo starts out by saying that all the techniques described in his book assume that your belay anchors are completely solid. Based on the first edition of Long's book, he recommends a rigging method called a "cordelette": a bigass loop of cord which you clip to all the gear in your anchor, then draw together in a big overhand knot and clip into.
Brutally simple¹, extensionless, equalised by construction, and only requiring a bigass sling, which is a good thing to have anyway: what's not to like?
He also describes a system called a "sliding X":
but almost all the illustrations in his book show cordelettes.
He also discusses the best material for making cordelettes: "5.5mm Blue Water Titan Cord is tested at 2200kg, while standard 7mm accessory cord tests at 900kg. Obviously, the 5.5mm Titan cord is better suited, though because of its slippery nature [it] should be tied with a double or triple fisherman's knot."
Obviously. It's common sense, right? 2200kg is clearly better than 900kg in this case, and the fact that the 5.5mm cord is lighter makes the choice even clearer.
Unfortunately, this is one of those cases where common sense can get you killed.
For the second edition of their book, Long and Gaines had an extensive series of lab tests done, and found that in terms of equalisation the cordelette is actually the poorest performing system tested, particularly (as in almost all real-world climbing situations) if the "arms" of the cordelette are of unequal length. The cordelette's performance becomes particularly awful (with a difference between arms averaging 4.5kN) when it's rigged with high-test cord.
But why could this be? Well, watch this suggestive (and terrifying) video of some tests performed by the Welsh gear firm DMM. Their commentary on the video is here.
tl;dw: Dyneema (basically the same stuff as the high-test cord mentioned earlier, and with similar equalisation performance) has very high static strength (the amount of weight it can hold up at rest), but much less give and stretch than ordinary nylon. This means, surprisingly enough, that it has lower dynamic strength: it's easier to break it by dropping a heavy weight (such as, say, a climber) onto it. Moreover, it transmits more force onto its attachment points. This lack of elasticity probably explains the poor equalisation, too. There's a more detailed investigation of various high-strength cords at Tom Moyer's site, which is clearly a site I'm going to have to devote some time to reading.
Long and Gaines now recommend a hybrid system called the "equalette", which combines most of the simplicity of the cordelette with the self-equalizing properties of the sliding X: the tests show that this outperforms both the older systems. For my part, I've always tied in to the anchor directly with the rope, as the DMM guy advises; this is more fiddly than tying in with a cordelette, and suffers from having a preferred direction of loading, but introduces more dynamicity into the system. Unfortunately, it's not one of the configurations that Long and Gaines had tested; looks like it's time for me to hit up Google Scholar.
How many people died because of this bad advice (which was and is widespread: I've unfairly singled Fasulo out, but it's repeated in many other books). Probably not many, and certainly fewer than were saved by the good advice in the books. As Long and Gaines repeatedly point out, anchor failure's a rare event, and anchor failure due to choice of rigging strategy is even rarer: it's much more important to ensure that your actual gear placements are solid. But probably one or two.
The real question is "why was this advice promulgated before the lab tests were done?" I'm guessing the answer is somewhere between "worl, s'obvious, innit? Why bother testing?" and "lack of money". As the sport expands, more money for testing should become available; and hopefully this kind of thing will convince the climbing community that while common sense may be enough in everyday life, it can be a woefully inadequate guide when lives are on the line.
This is, I think, a general principle. Reliance on common sense can get you killed. If lives are on the line, do science whenever possible.
¹ As legendary Himalayan climber Ed Viesturs points out in this fascinating interview, tiny errors can have fatal consequences in the mountains. Simple is good in mountaineering.
Note: as should be obvious, climbing and mountaineering are activities with a danger of personal injury or death, and participants in these activities should be aware of and accept these risks and be responsible for their own actions and involvement.