Very sorry Agent Smith, your path is one i have been on and found it was like on topic; but a distraction to me like watching the wrong hand of the magician when trying to crack consistent pivotal secrets of the functions. i don't take what you said as hate speech, nor intended to return such. Would almost expect views as you lay out, and even seen many more others that can't get even that far.
.
Lots said, will claim:
Any force (and rigidity) in the working knot must come from and go to somewhere; in balanced, accountable form. Math is the balance sheet language. cos:sin ratio is the wave form of change at deformity from linear, while carrying linear force input (ported in thru linear only device: SPart/s).
.
i think cos:sin, at least as a model/if not specific numerics, has a pivotal amount to do with all rope WORK>>especially visible enough to catch a glance of distilled from simplest forms
Everything must balance out in forces to how the finite Linear force input(in this model) is used.
cos:sin show the distribution ratio to the aligned dimensional axis vs. crossing dimensional axis(everything else); to a sum total of all expressed. The 2 dimensions are void of each other, points between are a hybrid mixture of both tho and to cos:sin ratio. As a function this is much greater than geometry, but perhaps best illuminated by geometry.
These divisions of force power against load and against host(seating). The seating in turn powers the friction, nips, grips. The seating to host for these functions from a focused linear(not dispersed radial)force type is what try to show. At deformities from linear the seating factor is way above nominal of simple linear in radial dispersion to host from focused linear input thru SPart. Bowline or (k)not.
.
Any direction of same angle pull is same: if 1 truck pulls the other home, the rope works harder than the pull service if the rope is at an angle.
But same load at same angle of pull is same tension force at any compass direction of pull; the forces don't care as much about direction (but rather dimension/s).
Hardline/fixed cos:sin positions is a people thing, not an organic force thing.
.
force direction is definitive to dimension claimed:i generally show systems as loaded; and the direction/type of input force is definitive to the logic.
The Bowline loaded as shown is linear force loaded; as like Hitch and Bend USAGE(not naming convention).
So, in this model, Bowline will follow the same usages of directional elements arc0,90,180 and their definitive given properties to define Bowline. It is radial force input types of binding tight as against swell potential, that is different handling somewhat. Bowline, as shown, has a definitive SPart that can only port simpler linear force into the magic arcs of real change.
.
pull/tension imparts a doubly key factor into rope/flexibles: ''tensioned rigidity" To me rope/flexibles tension carrys a double impact vs. a rigid under tension.
The Bowline, as others, host for nipping loop is truly not rigid/solid.
>>but becomes so on loading.
So the tension is not only the pull force, but also the rigidity factor, so i say 'tensioned rigidity' as force gives double duty hear.
If tried to 'nip' iron bar with soft wood crossing, iron would not be all that im-pressed;
but in reverse iron nip would im-press wood.
But the major rigidity of iron and wood are not from the tension, like they are in rope.
Under loading, proper ratios of rigidity reign thru the rope, where weren't before.
So we have greater tensioned rigidity parts and lesser tensioned rigidity parts of lacing when using the focused linear force type as input conversion to arcs. But not between the nips in binding.
Density of tensioned rigidity factor can be seen host is fatter, less densely compacted tensioned rigidity than smaller rope of same tension and material grip around host AND dent host !! Not just trapping steel bar with soft wooden bar, but denting of steel denting soft wood as traps it too.
.
Bigger than geometry:The cosine curve/waveform repeats in many things, as an Universal organic pattern of change from 100%-0%; therefore all possible values. Geometry is a most visible, tangible, enumerable etc. gateway to see these Universal workings of change over full range of 0-100% expression of a potential; and then too at same the the flip side. Both sides of a flipping coin at once. A sine wave can show the drift extremes and the middle, organically balanced, aligned cos center of pendulum swing, piston travel, spring recoil etc.; to show all values of aligned and non-aligned considerations. As a child in swing can feel the cosine_0 of motion range as breathless stall before return swing. The bottom of swing is the organically aligned cos_1 of the swing, sines to the side(cos_0). Sine Wave of extremes of cos_0 to sides with center of organic balanced cosine_1, as a waveform xTension, xDistance(etc.).
.
benchmark cos and drift of sinei take the linear force line input direction, the instigator, as full benchmark cos axis to trace; and 90 from as full sin.
A dynamic, not fixed/static/'stagnant' usage. cos:sin number scales are reciprocal, so can switch names and is all the same. But perpendicular references are in different dimensions is absolute. To me, cos is measure of alignment and sine measure of deflection to ultimate full crossing of a different dimension. A 1D linear benchmark(cos) that drifts into 2D space(with sine deflection from cos). Ancient's system is so solid, can use it again and take the examined 2D reference now as benchmark cos, and drift from into 3rd D as sine!!!
.

.

.
In different dimensions to me means 1 dimension will NOT effect each other's total 'rigid displacement against' (key terms/factors) distance (or it's reciprocal of force) total quantity/s in it's own dimension. Just as a pure North path (as N/S dimension of travel) would not have any effect on how far E or W(from benchmark); only a reduction to further negative S value. Even if wind blows leaf pure horizontal, it still has the same journey vertically down, just spread over larger area.
.
Peace