Wouldn't you agree that if you had a 10 point harness, that if a load L is applied anywhere in the system that at any given point the load on any one mounting point or loop/pulley is not greater than the initial applied force?
I disagree, and I'll provide an example using an everyday three-point seatbelt system.
Note: A pulley can be used to both change the direction of tension and as a force multiplier. A belt guide, not belt endpoint, is basically a pulley with some friction.
Sit in the seat and pull the seatbelt over your chest. The tongue/male blade slides along the length of the belt via a guide that allows for ease of use and adjustability. The guide is in its simplest form a pulley. Let's ignore the friction component for this example since it is not needed to illustrate the basic reactions. The belt is now attached at two points.
One one end of the belt is the retractor mechanism, the other a hard mounting point.
Finish putting the seatbelt on by inserting the tongue into the latch. There are now three mounting points.
Pull up on the section of belt that is going across your chest so that the belt across your lap increases in tension. Now you have tension in the shoulder and lap belt passing through the guide in the tongue blade. If you pull on the belt as described above with 10#, then the lap belt also must have a 10# tension.
For ease of understanding assume that the shoulder and lap belt sections are parallel to each other as they enter the guide on the tongue blade. The tongue blade must therefore have a reaction of 20#. Hence, the reaction at the mounting point is 20#. The belt only has 10# of tension, but the guide bar, tongue, buckle, all the way down and including the bolt/floorboard must be designed to withstand a 20# load. This is a worst-case scenario.
The belts as they enter the guide on the tongue will typically not be parallel when inserted into the latch, but more than enough to create a reaction load far greater than the magnitude of the belt tension. At 90* to one another, as noted in the previous reply, is sqrt2.
I am just saying that you can't gain force in a closed setup. if the force on the belt at one point is "L", then the force on any one mounting location on a multiple mounting location setup is going to be less than L.
That is one of the primary reasons for using pulleys. Force multiplication. It's not what we want in some cases, but it's also how a single man can lift something dozens of times his own weight. It is also why the reaction at the tongue blade guide is significantly higher than the belt tension alone.
Some math:
A rope/belt is pulled horizontally at a tension T. In the middle of the belt (F) a frictionless pulley of negligible diameter is placed below the belt. Find the angle theta (0) to the horizontal where the vertical reaction (Fr) at F is the same as T. Assume that both ends of the belt are lowered an equal amount to the horizontal. (A drawing is worth a thousand words...)
Assume the downward direction is negative, upward positive. No horizontal movement occurs since both ends of the belt are dropped equally, hence the horizontal components remain equal and opposite.
Since we want to find the point where the belt tension is equal to the reaction Fr, then Fr=T.
For solving purposes, Fr=T=1
Using statics:
-Tsin(theta) - Tsin(theta) + Fr = 0
-2Tsin(theta) + Fr = 0
Tsin(theta) = Fr/2
sin(theta) = 1/2 {Fr = T = 1}
theta = 30*
Hence, whenever a minor/inside angle of less than 120* (180 - 2x30*) occurs then the reaction at the guide reaction point will be greater than the belt tension.
This is all a very simplistic engineering viewpoint. It does not take into account a multitude of other factors that come into play.
The reaction at the other end of the belt (opposite the latch) will be 10#, assuming there is no friction between the belt and your clothes.
Your never-greater-than-the-belt-tension reaction only works on a two-point lap belt type setup.