I have a theory too wrote:
As an example, we don't inherently know how to ride a bike, we have to learn it. Once we learn how, we don't explicitly think 'turn left, turn right, lean left, lean right', this is relegated to a subconscious level and takes over the process. In this way, we've trained ourselves to ride a bike and riding itself becomes easier as we don't have to try to balance. This doesn't apply in running. We know how to run, that part is easy. What we lack as newbies is the conditioning, and it's the physical stress part that's difficult.
Funny example. I have a one year old daughter. She doesn't know how to run yet (or even walk), but she will have to learn it, until it is relegated to a subconscious level and takes over the process.
I have a six year old son who can ride a bike. Yet he cannot climb the Mont Ventoux. What he lacks as a newbie is the conditioning -- it's the physical stress part that's difficult.
How is running and cycling different again?
Before I get into my other questions, let me say thanks for your long summary, and that in large part, I agree with you, or at least understand what you're trying to say. I'm not sure I agree with your dismissing CG as a wrapper though.
After reading everyone's criticisms so far, as a general comment, I still don't understand what people fear about the Central Governor model.
Perhaps in part, it's the name. Calling it the Central Governor makes people think it's some kind of puppet master who has final control, and this makes some insecure, as if they were not in control of their voluntary functions. Maybe if the name were "Autonomous Regulator" or "Developed experience with bio-feedback mechanisms", the idea might gain more acceptance.
To establish the context, my understanding is that this CG model is intended to replace an aging, 90 year old, physical limitation model, where exercise is limited by oxygen, and lactic acid makes muscles tired, and the brain doesn't play a role in limiting performance. No one seems to be actually arguing for the old Hill model, beyond defending that pseudo-LT and pseudo-VO2max are useful concepts in training, even if they are not real (a point I would argue for and agree with). So the other main arguments seem to be that CG has not been shown to be the right model, so we can't pretend it's a fact yet. Or simpler explanations exist, as you've suggested, and observations can be simply explained by experience or cognitive learning.
You write that "learning works at many levels in our minds, not just the top 'conscious' level. I consider this something already known and don't see the need to apply a CG wrapper around it". Maybe at the track it is known, but, as I understand it, in exercise physiology, it is not already known. Is there another model which incorporates this "multi-level" learning, besides CG. Not because CG is the best, or even good, but simply because scientists haven't really started looking much in this area. Is anyone looking at how learning and experience impacts racing performance? Is anyone else looking at how heat or altitude triggers pacing adjustments? Perhaps there are better models or theories or hypotheses, but I just wonder if anyone is looking at other models, or if proposing any alternatives suffer the same fatal flaws that many accuse the CG model: no one has yet shown these things to be true, and therefore these simpler CG alternatives are mere speculations or hypotheses.
In the example of the runner who starts the marathon at 10K pace, there is no inconsistency with respect to overriding the CG. If you think of the CG as keeping the body out of danger, then running 10K at 10K pace should not put the body in danger, so there is no overriding of the CG. However, the experienced runner knows almost instantly what the optimal marathon pace is, for any given altitude and temperature. A runner can override this optimal marathon pace all he wants, until he puts his body in danger. Perhaps you have a point that this "experience" component should not be part of CG, and maybe one day it won't be part of CG, or CG's successor.
I'm not sure I understand your problem with the "anticipatory" aspects of CG either, unless it's a question of perspective. For example, Dr. Noakes gave the description of how starting an exercise with high glycogen or low glycogen ends up with the same level of glycogen at the end. Now you could say the "CG" was reacting to the low blood-sugar level, in anticipation of depleting the glycogen stores. Is there something wrong with that statement? Are you dismissing the anticipatory aspect because the "CG" is just reacting to low blood sugar (or low oxygen, high temperature, etc.)?