And here it is, an answer by Jesus Dapena. He himself calls it a 'a very looooooooooong email with my response'. Here is the text:
This is a funny position for me to be in, because I have never been a referee before. But I do officiate in track meets, so maybe that counts a little!
I believe that both Weia and Glen are right in most things, and wrong in a few things. So I have to declare a tie in the battle. Solomon would be proud of me! No, seriously, as I see it, there is a mixture of a lot of right and a little bit of wrong in how you both are viewing the jump. Overall, your trains of thought are both pretty good.
First, some of my terminology, so we will understand each other more clearly: There is a "takeoff phase", which lasts from the "instant of touchdown" (or "plant") to the "instant of takeoff".
Glen says (a) that you need to use a gently arched position throughout the flight and (b) that it is unnecessary to lift your feet in the late stages of the bar clearance. I don't agree with either of these statements. A high jump specialist needs to arch quite a bit, including a strong bending of the knees (as if you wanted to kick the bar from below with your heels), and then with just the right timing you need to "un-arch", meaning that you will flex at the waist and you will kick up your feet. By arching a lot, the pelvis will reach a greater height (the legs go down, the head and shoulders go down, and the hips go up; the center of mass [c.m.] continues in its predetermined path). But if you arch this much and then stay arched you will surely dislodge the bar with your calves, because nobody has enough angular momentum to rotate quickly enough to avoid such a problem. So after the pelvis has crossed the bar the athlete will need to un-arch, or else the calves will dislodge the bar. The un-arching needs to be executed with very good timing. If you un-arch too early, you will "sit" on the bar, and your butt will knock the bar down, as Glen says; if you un-arch too late, your calves will hit the bar. A rough rule of thumb is that you should start the un-arching when the bar is at about 1/3 of the distance down the thigh from your hip joints toward your knees. Glen's proposed technique (mild arching and little leg kick) will also sort-of work, but I would recommend it only for "safety" jumps in which you know you will have a lot of space between you and the bar, and in which you do not want to take any risks of hitting the bar through errors in the timing of an un-arch. But when you are going for the big heights you have to do the big arch, followed by the un-arching with good timing. There are some exceptions to this general rule: (a) decathletes may be happy with the added consistency of the modest arching proposed by Glen; also (b) some high jump specialists who happen to execute their run-up and takeoff in such a way that they are left with a very large amount of leftover horizontal speed at the end of the takeoff may have to stick to a weak arching technique because they simply won't have enough time to arch and then to un-arch in time, because there is very little time from the instant when the shoulders cross the bar to the instant when the hips clear the bar and to the instant when the feet clear the bar, because you are traveling so fast in the horizontal direction.
Glen talks about a "lateral rotation" about a horizontal axis parallel to the bar and a "somersaulting rotation" about a horizontal axis perpendicular to the bar. This is perfectly correct, it is one way of describing the rotations that make the shoulders go down and the legs go up. And I use axes like these for part of my numerical calculations of angular momentum. However, I prefer a different way of breaking down the angular momentum into components, because I think it is more helpful for understanding the causes of the problems that can occur in the generation of angular momentum during the takeoff phase. So this is not a matter of right versus wrong, but a matter of useful versus less useful. Like Glen, I use two horizontal axes for my descriptions of angular momentum, but my axes are different from his. While his axes are described relative to the bar (parallel to the bar and perpendicular to the bar), the axes I use are described relative to the final direction of the run-up. What is the final direction of the run-up? At the start of the run-up, the athlete is running perpendicular to the bar; once the curve starts, the direction of motion of the athlete (of his c.m., to be precise) gets progressively less and less perpendicular to the bar and more and more parallel to the bar. Therefore, immediately before the takeoff foot is planted on the ground to start the takeoff phase at the end of the run-up, the athlete will be running at an oblique angle relative to the bar, somewhere intermediate between perpendicular to the bar and parallel to the bar. I call this angle the "final direction of the run-up", and it is about 35 or 40 degrees (if we use a terminology that says that parallel to the bar is 0 degrees, and perpendicular to the bar is 90 degrees). The two components of angular momentum that I define are the lateral somersaulting angular momentum and the forward somersaulting angular momentum.
The lateral somersaulting angular momentum produces rotation about a horizontal axis aligned with the final direction of the run-up. If you were running behind the high jumper, you would see this as a "cartwheeling rotation" toward the right. If there were only lateral somersaulting angular momentum and no forward somersaulting angular momentum, the athlete would end up horizontal at the peak of the jump, but at an angle to the bar (when viewed from overhead): The head would be closer to the right standard and the feet closer to the left standard.
The forward somersaulting angular momentum produces rotation about a horizontal axis perpendicular to the final direction of the run-up. Imagine a gymnast approaching the bar with a straight run-up at a 35 or 40 degree angle to the bar, and then making a purely forward somersault (as you would do in gymnastics). If there were only forward somersaulting angular momentum and no lateral somersaulting angular momentum, the athlete would end up again horizontal at the peak of the jump, but at an angle to the bar (when viewed from overhead): In this case, the head would be closer to the left standard and the feet closer to the right standard.
What high jumpers need to do is to generate BOTH lateral and forward somersaulting angular momentum. This will amount to a "total" somersaulting angular momentum, which will normally make the athlete be horizontal at the peak of the jump, perpendicular to the bar in the view from overhead, and it will be a faster rotation than either the laterel or forward somersaulting angular momentum alone would achieve. This "happy" result occurs when the lateral and forward components of angular momentum are in the precise correct proportions. (This perfect combination is the rotation that Glen calls the "lateral rotation".) But if there is too much lateral somersaulting angular momentum, the head will end up closer to the right standard; if there is too much forward somersaulting angular momentum, the head will end up closer to the left standard.
(The "happy" total somersaulting angular momentum is what Glen calls the "lateral" rotation. What Glen calls the "somersaulting angular momentum" is what happens when either "my" forward or "my" lateral components of angular momentum are too large. In such case, the total somersault rotation is not perfectly about an axis parallel to the bar, so that there is an additional component of angular momentum about an axis perpendicular to the bar.)
In addition to the somersault angular momentun components that we have seen until now (lateral somersaulting angular momentum, forward somersaulting angular momentum, and their sum the total somersaulting angular momentum), the high jumper also has a "twist" angular momentum, which produces a counterclockwise rotation about the vertical axis. Rotation about this axis is very easy to achieve. It is generally produced by throwing the right leg forward and them diagonally toward the left at the end of the fakeoff. At the end of the takeoff, the thigh will normally be parallel to the bar; it normally does not go 20-50 degrees more counterclockwise than that (as Weia believes). If may FEEL that way to the jumper, but it does not really happen that way. Glen has a very good description of this:
> I don't think that leg swing (meaning lead leg drive) has anything to do with rotation about
> the bar - or at least it doesn't for the method of approach and takeoff I teach. However, I
> do think it should be the primary method of inducing rotation about the vertical axis during
> takeoff. I think the knee should be driven up to the level of the hip (level upper leg) and
> slightly across the body so the thigh ends up roughly parallel to the bar.
A possible exception is when an athlete uses a "running" arm action (like the one used by Fosbury himself, although it is pretty rare today). The running arm action is often accompanied by an increased leftward losition of the lead leg, i.e., more across the body, and beyond the parallel to the bar. This may be an action/reaction thing: Seen from above, the running arm action is clockwise, and the leg drive is counterclockwise.
The right leg is an important factor in the generation of "twist" angular momentum, but it has little to do with the generation of the somersault angular momentum components. In fact, the little that it has to do is a disadvantageous effect. The somersulting components of angular momentum are produced through active rotations made by the athlete's whole body during the takeoff phase: The athlete goes from a position tilted to the left at the start of the takeoff phase to a vertical position (as seen from behind) or a little beyond the vertical at the end of the takeoff phase --this generates the lateral somersaulting angular momentum; The athlete also goes from a backward-leaning position at the start of the takeoff phase to a vertical position at the end of the takeoff phase (vertical as seen from the side of the athlete) --this generates the forward somersaulting angular momentum. The lifting of the lead leg during the takeoff phase, while useful for the generation of lift for the jump, actually interferes some with the generation of forward somersaulting angular momentum. Seen from the left side, the high jumper needs to generate a counterclockwise rotation (the forward somersaulting angular momentum). But if the right leg swings hard up, that is a clockwise rotation, and interferes with the generation of forward somersaulting angular momentum. So you end up with a smallish amount of forward somersaulting angular momentum at the end of the takeoff, and therefore with a smallish total amount of somersaulting angular momentum, and in addition the proportions between the forward and lateral components becomes too uneven in favor of the lateral somersaulting angular momentum. As a result of all this, (1) the athlete finds it hard to somersault enough, and in addition (2) the head will be closer to the right standard than the feet. (1) is by far the larger problem. Sotomayor is a clear example of this problem: very mediocre bar clearance, although he was phenomenal in the generation of lift --as everybody knows!
Glen says:
> The disadvantages of the straight (or straightened curve) approach are the loss of height due
> to a laterally angled takeoff and the increase in likelyhood of upper-body contact with the
> bar on the way up (also due to a laterally angled takeoff.
That is a very true statement. Just one comment: The curve serves (among some other things) to tilt the athlete toward the left, so that he/she can rotate toward the right during the takeoff phase (while he generates "my" lateral somersaulting angular momentum), and still don't go much beyoud the vertical at the end of the takeoff phase. So the curve helps only the lateral component of somersaulting angular momentum. The forward ("my" forward) component of somersaulting angular momentum has nothing to do with the curve; it relies on the athlete acquiring a backward tilt in the last step of the run-up, and then rotating forward to the vertical by the end of the takeoff phase.
Once you have left the ground your angular momentum is fixed, it can't be changed until you land on the pit. But you can still alter (within certain limits) how fast you somersault. If you make your body be more compact in the view along the bar, you will somersault faster for any given amount of somersaulting angular momentum that you may have. I believe this is what Weia was getting at when she talked of "stretching of the hip". One way to make the body be more compact in the view along the bar is to bring your arms by your side as you clear the bar; don't have them extended away far beyond the plane of the bar and the standards. But the best way to make the body be compact in the view along the bar is to flex the knees a lot during the bar clearance. By this I do NOT mean a fetus-like position, but a markedly arched position, with the knees very flexed, in such a way that it will seem that you are trying to kick the bar up with your heels. Of course, you can't hold this position forever, or you will dislodge the bar with your calves. So you will need to UN-arch with good timing.
Glen is right when he says that the action of the right leg during the takeoff phase affects the twist rotation of the body about the vertical axis. Too little drive across the body, and at the peak of the jump the athlete will be tilted with his right hip lower than the left hip; too much drive across the body, and at the peak of the jump the athlete will be tilted with his left hip lower than the right hip.
A separate story is that this unwanted left/right tilt of the pelvis at the peak of the jump is not only the result of the right leg drive during the takeoff, but also of the (dis)proportion between the forward and lateral components of somersaulting angular momentum: Too much forward component and the right hip will be low at the peak; too little forward component (or too much lateral), and the left hip will be low at the peak. But this is a rather complex issue to explain, so I'll be merciful and leave it for another occasion!
So Weia is very right when she says:
> You often see jumpers with one hip higher above the bar than the other hip. This will cost them
> some millimeters, up to 15 maybe. That can be the difference between gold or silver, so it is
> to be avoided. I see this rotation as a result of a mismatch between forward somersault and
> somersault sideways.
Weia says ...
> All upward momentum is produced by the takeoff leg. In the past I thought that a fast upward
> swing of the swing leg and/or the arms added upward velocity, but in newtonian terms that is
> not correct. What happens is that the upward movement of a leg or arm is countered by a downward
> movement of the rest of the body. This means that the pressure to the ground becomes bigger,
> that leads to a bigger ground reaction force and so more upward momentumm. But only if the
> takeoff leg is strong enough to withstand the pressure, so it still is that leg that produces
> the upward momentum.
This is a very good description. There are 3 factors that contribute to compress the takeoff leg against the ground (which is a good thing): fast run-up, low position at the end of the run-up, strong upward actions of the free limbs (arms and right leg). Too little of these factors, and the jump won't be as high as it could have been. But there is a limit: If you go too fast or get too low or swing the arms and lead leg too hard, the takeoff leg will buckle (collapse), either completely or partly, and the result will be a very bad jump. So these factors are good, but there is a limit, an optimum, beyond which the jump will become bad. Stronger athletes can handle faster and lower run-ups, and stronger actions of the free limbs without buckling, while weaker athletes should use more moderate combinations.
Jesus Dapena