What is the basis for Stu's estimate of lower strength of the soft shackle with a stevedore knot as the stopper? Just conservatism?

Attractive and symmetric as the diamond knot is, I can see how it might be better for retained strength, but still not great.

Suggestions or experiments of of a stopper knot that retains more of the strength of the line?

I'm trying easy-to-make zeppelin / Rosendahl bends with doubled initial loops and two more passes of the ends through those loops.* That bulks up the knot, and to the point, enlarges all the bending radiuses, which should strain the cord less and retain more strength. But I have no way to test the result.

*You always see illustrations with the two standing ends going in opposite directions. For the stopper use, the ropes in the illustration would be on the same side