1. Plenty of instructors express doubts about the effects of hauling the line during the cast; however lots of discussions on basic physics and some video experiments (vimeo.com/album/1532129) seem to have had an effect on how some of them address this issue lately.

Nonetheless a last objection remains: They say that, although hauling obviously does load the rod, mentioning it is misleading, since the main goal of hauling is to give more speed to the line via the line hand.
And this is very true: rod loading is only a byproduct of the haul; but why point out this issue only when it comes to hauling? In fact rod loading is always a byproduct of the casting process.

For years flycasting has been summarized in sentences like “loading the rod and then unloading it to propel the line”. This mantra, repeated so many times, led to the idea that every action by the caster that results in rod loading is good just due to that; logically, from that we should deduct that our casting problems come from failing to load the rod.

This is totally misleading because the main use of loading/unloading the rod is not to propel the line whatsoever. The function of rod loading is to allow us to apply force in the direction of the target along a longer distance, so we can transfer more energy to the line.

And you might be wondering: What does all of this have to do with the title of this experiment? Is this guy getting mad? Well, not totally… yet. :-D

The difficulties of practicing roll casts on grass have been traditionally explained by "the grass doesn't grip the line like the water does, so the rod doesn't load, making the cast more difficult". And what I had in mind was: if supporting that hauling loads the rod (something that, in fact, does) is considered misleading, what to say about defending that the anchor in a roll cast loads the rod, if, in fact, it doesn't?
This has encouraged me to edit some new footage to analyze the origin of the problems with the roll cast.

This previous material might be of interest also:

And also this one:
vimeo.com/album/1529848

I might be wrong in my analysis, so any feedback will be very welcome.

Edit 25/8/011:

I have found a comment:
sexyloops.co.uk/cgi-bin/theboard_07/ikonboard.cgi?act=ST;f=24;t=13718
It says that "the anchor loads the rod" isn't a very common concept among instructors.
Well, I have read that statement so many times that I think just the opposite: "the anchor loads the rod" has been the most common explanation of the role of the anchor in roll and spey casts.
Just one example taken from page 9 of the book Two-Handed Fly Casting by Al Burh (member of the FFF's Board of Governors and chairman of the Two-Handed Casting Instructor Program):
"The line that is allowed to anchor or rest on the surface will in turn be the resistance or load for the forward cast."

2. Traditional fly casting instruction states that the line follows the rod tip. This video:
vimeo.com/aitorc/does-the-line-follow-the-rod-tip
shows how that idea is just a rough approximation to what actually happens.
One example of the limited value of that statement is bad tracking: if the back and the forward casts are separated by less than 180º in the horizontal plane strange things happen on the tip of the line. Why is that possible if "the line always follows the rod tip"?

In this new video example we can clearly see what happens when we have a faulty tracking:
1.- Since only part of the line moves in the direction of the rod tip motion a wave appears in the rod leg of the loop; that wave travels down the line.
2.-The tip of the line is put in motion in a different direction, and as Newton taught us, that length of line tries to keep moving in that direction due to its inertia. At a given time, the line and the leader are positioned 90º from each other. That's the explanation for the curve cast presentation that sometimes results from these faulty casts.

The same phenomenon (though in a more vertically oriented plane) is behind the problem of the "dangling end" that appears when casting medium to long lengths of line:
vimeo.com/aitorc/dangling-end-explained

3. A sequel of this previous video:

This is how I understand it by now. Most probably I am missing something... or getting it wrong.
Feedback is very welcome.

Edit:
Due to my very limited knowledge of physics what I consider to be a wave maybe isn't a real wave in a technical sense.
Anyway, wave or simple bend in the line, it is what explains the source of the Dangling End on long carries and other similar issues due to tracking faults.

4. "When moving the rod forward parallel to the surface the rod tip will curve to the left, causing the leader and line end to follow the same curved path."

"Rebound is essential for making an overpowered curve."

We have read statements like those quoted above a lot of times. Is this what really happens?
This experiment shows that an overpowered curve in the end of the "line" doesn't need any counterflex or rebound in the rod to be made.
In fact the only condition needed is that when the loop gets straight there is some energy left in the line; the more energy left the more pronounced the effect.