Bowline Strength Test - Conventional Wisdom Bites the Dust

My November 13 post was about the "correct" way to tie a bowline. Searching the web, I came across some comparison testing of the correct and incorrect methods (http://user.xmission.com/~tmoyer/testing/pull_tests_11_98.html). The testing was by a member of the Salt Lake City Sheriff's Office, which does a lot of mountain rescue. The protocol looked a little unorthodox but sound (see below). Here is the result of the bowline test:

            11/23 Test #14
              Pull a bowline knot to failure. A bowline was tied in each end of a piece of
              new 11 mm Blue Water rope. One knot was tied "correctly". The other was
              tied "incorrectly", with the tail of the rope outside the loop formed by the
              bowline. The load was applied between the two bowline knots, on a single
              strand of rope.
              Result: Material failure at the "correctly" tied knot at 4840 lbs.

The caveat I will add is that these knots were presumably both drawn up tight before they took a load. Just maybe the correctly tied knot is more stable and less likely to trip while drawing tight, a consideration if - but only if - the line is so big and stiff that it cannot be drawn up by hand - not a common scenario for the yachtsman. Also, with the tail inside the knot the correct bowline is a bit less likely to hang up, if that is an issue.

Here is the testing protocol:

"We used a vehicle winch on a Hummvee to apply forces. A second Hummvee was initially used as an anchor. However, with an end-to-end pull, and with all four wheels locked, we were able to drag both vehicles across the concrete floor with 5000 lbs force. (good number to know if you use vehicles as anchors.) We ended up anchoring one vehicle to a tank (yes - really) and the other to eyebolts mounted in the wall. A Sensotec load cell was used to measure forces. It is calibrated internally with a shunt resistor."