You can't analyze a suspended vehicle this way. How can you say anything clear about a dynamic system like a bike and rider by analyzing it as if it's at rest?
As Archimedes pointed out a couple thousand years ago, adding input at multiple points along a mechanically joined thing acts exactly the same as if all the input happens at one point right through the center of gravity. Common sense states that if you take your hands off the bike, the pedals will still put a bunch of pressure right on the front wheel, and the way you unweight a wheel isn't via your hands on/off or your feet on/off, but by adding torques to the bike and also by shifting your body's weight to deliver momentum, unweight, etc. by moving the momentum of your own center of mass.
The way that this video's analysis works, it commits a fundamental mistake where it assumes that the hands put the weight on the front wheel and the legs manage the back wheel, but there's no overlap and perfect division of duties. It's further messed up because this analysis treats the bike as if it's a 2D structure, which it isn't. Bikes have a left and right side, and the hands and pedals are not on the mid-line.
Bottom line, you cannot analyze a dynamic engineering system as if it's a static engineering system or you start making nonsense.
Imagine if someone analyzed a suspension linkage at full rest or maximum extension, without taking into account that the system moves, there are valves, there are fluids, there are pressures and that its work is delivered through a bunch of moving parts like linkages and frames. Makes sense that that analysis would be super flawed, right? This is why we talk about the kinematics of the whole moving system.
If you don't think I'm making any sense, here's an experiment for you: put a bathroom scale under your front wheel and stand still. Now see what you have to do to make the weight go away. If your can't unweight your front wheel by any static resting position, how valid is this whole method of analysis?