In this article we dig a little deeper into the biomechanical factors in the positive aspects for jumps.  In part this was motivated by contemplation of how a machine vision system might quantitatively evaluate jump executions using numerical metrics.

Big Giant Disclaimer - This is not about how the positive aspects should be, or are, judged under IJS; or even whether these are the correct aspects that should be considered in judging.  Rather it is an exploration of the biomechanical concepts included in the positive aspect bullet points.

Of the six positive aspects for jump, some lend themselves to a biomechanical analysis while others do not.

1) Very good height and very good length (of all jumps in a combo or sequence)

What might be considered very good height we can categorize based on the heights of jumps found in studies of countermovement jump for elite athletes across multiple sports and the heights of jumps in competition.  These studies consistently show elite athletes perform countermovement jumps in the range from 10 to 33 inches.  Large statistical studies for skaters on ice do not exists (as far as we have determined) but in our own more limited analysis of jumps, the highest jump we have measured is near 32 inches, and typical jump heights on ice appear to be in the range of 12 to 16 inches.

With the above values for heights, as well as the standard deviation intervals measured for countermovement jumps, we offer the following table relating numerical values to descriptive words for jump height.

Column 1: Description of the jump height
Column 2: Range of jump height
Column 3: Range of time in the air
Column 4: Range of rotations completed in the air for an average rotation rate in the air of 5 per second (an average rate that is commonly achieved by skaters of most ages, other than adult competitors).

The length traveled by a jump in the air is determined by the time in the air and the speed at the instant of take-off.  As a starting point, we can then say for the typical takeoff speed of jumps, to have a very good length you need a very good height.  On that basis the following table relates numerical values to jump length description.

Column 1: Description of the jump length in the air
Column 2: Range of length in the air for a common takeoff speed of 10 ft/sec

Note that the length table does not guarantee a jump of very good height will have a very good length.  A jump of lesser height taking off at a higher speed can achieve a very good length;  and of course, a jump of very good or excellent height might not have a very good length if the take-off speed is low, as in when some skaters come to a near stop at the instant of takeoff.

One visual characteristic of jumps with very good height AND very good speed is that the parabolic arc of the flight has a pleasing shape, neither tall and skinny, nor low and broad.  Other than for the small and very small jumps, the length to height ratio for that pleasing shape is roughly in the range 3.5 to 4.5.

A ratio of 4:1 (length to height) results from a take-off where the horizontal speed is twice the vertical speed at the instant of take-off.  Note that jumps that are neither high nor long can also have a pleasing arc of flight, so long as the speed ratio is maintained, they just aren't very big.

While most jumps can have very good length if they take off fast enough, no jump up through about triple Salchow has very good height when held to the above absolute standard, given that jumps through triple Salchow are generally completed successfully with only 16-20 inches of height, and for singles and double much less.  This is due to the way skaters control the number of rotations they fit into a jump.

Skaters primarily control the number of rotations completed in a jump by controlling the time in the air through the height of the jump.  That is, as the number of rotations in a jump increases, the average rotation rate shows only minor variation, while the height of the jump increases significantly to give more time in the air.  From singles through quads a good rule of thumb is that the average rotation rate will be around 5 rotations per second (+ perhaps 25%) with the height of the jump ranging from well less than 12 inches for a single through 32 inches for a quad, roughly a factor of 3.

As an example, most double Axels for skaters of all levels have about 0.5 seconds in the air and an average rotation rate of 5 per second.  You never see double Axel jumps with 0.8 seconds in the air and a rotation rate of 3 per second.  You might, however, see one with a bit less than 0.5 seconds in the air and a bit more than 5 rotations a second, but not by much.

2) Good take-off and landing

It could be argued that any take-off that gets the job done (a fully rotated clean jump) is a good takeoff, and any landing that returns the skater to the ice in a controlled secure fashion is a good landing.  Additional aspects of good takeoffs and landings is the amount of the rotation executed on the ice.

During the takeoff a skater develops a vertical force into the ice and a torque around the rotation axis of the skater. The ice exerts a reaction force on the skater that propels the skater into the air and sets the skater rotating (thank you very much to Newton's third law of motion).  This occurs over a time interval of approximately one-half second.  During that time the skater must rotate some small amount on the ice.

At the landing the skater again exerts a force and torque into the ice, and the ice responds to stop the vertical decent on the skater and slow the rotation.  During this time interval the skater must rotate some (hopefully) small amount on the ice.

There is no definition for the amount the skater can rotate on the ice at the takeoff and landing to be considered fully rotated.  However, from empirical observations we know that the amount of rotation on the ice for clean jumps is typically 1/4-1/3 rotation each on takeoff and landing, and if this amount of rotation is normal, that implies "good" would mean better than that, or less than 1/4 missing rotation.  On that basis one could apply the following adjectives to the missing rotation on takeoff and landing:

 Excellent - < 1/8
Good - 1/8 to < 1/4
Moderate - 1/4 to < 1/3
Poor - 1/3 to < 1/2
Under-rotated or worse - 1/2 or greater

Other factors necessary for a takeoff or landing to be considered good from a mechanical perspective are:

Takeoff on one foot with a clean correct takeoff edge.
No sudden loss of speed immediately prior to takeoff.

Landing on one foot with a clean correct landing edge.
Glide out of the jump with as much horizontal speed as the entry into the jump.
Control of the body in stopping the rotation of the jump.

The body position at the takeoff and landing is omitted here as there is a separate bullet for body position.  (No. 5)

3) Effortless throughout (including rhythm in jump combination)

For biomechanical analysis, one might ask, what exactly is the physical quantity "effort" that needs to be missing?  What exactly is "effort" and how does one measure it?

Some people use effort as an imprecise term for work done to achieve a task.  Others sometimes use it as a substitute term for force.  Both of these uses miss the mark for this bullet as all jumps require considerable force, torque and work and are thus never effortless.

Another common definition of the term effort is "a vigorous or determined attempt."  Does that mean jumps should lack a vigorous or determined attempt?  Not likely, as all successful jumps require a vigorous or determined attempt to generate the forces and torques need to achieve the height and rotation rate required for success.  So using that definition, all jumps are again never effortless.

 So what exactly is the biomechanical concept of effortless for jumps?

A clue to that lies in the term generally used for jumps that are not described as effortless, which is "labored."  While all jumps are never physically effortless, they can appear effortless or they can appear labored.  Skaters can make a jump look easy or like an ordeal.  This being the case, one can ask whether this bullet is anything more than the subjective impression of the ease with which a jump is executed, and question whether it is a measurable biomechanical concept.

 Perhaps the phrase "including rhythm in jump combinations" offers a saving hint.  This implies that the appearance of effortless is achieved by the smoothness, continuity and coordination of the motions, rhythm and timing of the execution.  This is something that is certainly measurable, though a complex and messy thing to do.

Not having a quantifiable metric to delve into effortless further we will have to leave it at that for now.

4) Steps before the jump, unexpected or creative entry

Either the steps are there or they are not.  Whether the entry is unexpected or creative is the subjective opinion of the viewer, and not a biomechanical concept.

5) Very good body position from take-off to landing

Putting aside the aesthetics of what a body position should look like, the challenge is to come up with a metric that would numerically quantify what is an excellent vs. a poor air position.  Part of the challenge is that there are many variations in air position that all get the job done.  The question then, is what is common among those variations that would result in a similar numerical metric.

To do this we come at it by asking the question, what is the main mechanical purpose for having a very good (or better) body position?  Our answer is, to minimize the moment of inertia of the skater to maximize the rotation rate in the air.  Thus, we would claim any body position that minimizes the moment of inertia of the skater is an excellent air position.

Given the level of physical skill required and the minute margin for error, we would claim that any clean fully rotated quad always has excellent body position from take off to landing.  For the most difficult triples (loop, flip, Lutz and Axel) the body position must be very good or excellent for clean fully rotated executions.  For singles through triple Salchow, jumps, our empirical observation is these jump might have excellent body positions or significantly less and still be clean and fully rotated.  That is, one can complete the less difficult jumps cleanly and fully rotated with flawed air position.

Rolling all of the above together, the following table captures our initial thoughts on a numerical characterization of air positions.  The numbers should be taken with a huge grain of salt as they are based on only a small number of jump examples.

Column 1: Description of the body position in the air
Column 2: Best moment of inertia of the skater during the execution
Column 3: Common position defects

(*)  When arms overhead variation is used, the flaws for the arms involved with those positions include how far the upper arms and elbow are from the rotation axis and whether the forearms intersect the rotation axis.

Not captured in any of this discussion is how long it takes the skater to get into the best position in the air nor how long they hold the best position in the air.  So the above should be taken to refer to the best position reached in the air during the execution regardless of how long it take to achieve that position and how long they hold it.

In addition to achieving the minimum moment of inertia, there is an additional "there-or-not" condition, that the rotation axis of the jump must be oriented vertically or nearly so.  When the rotation axis departs too much from vertical, jumps become nearly impossible to control on the landing and such a jump would not be considered to have a very good body position for the landing.

[Landing with the rotation axis far from vertical results in the body of the skater precessing like a top.]

Further, there are also the takeoff and landing positions to consider.  These are also there-or-not in nature from our perspective; that is, they are either very good or they are not, without shades of gray.

For the takeoff, the mechanical question is, does the takeoff position aid the rotation of the jump or inhibit it?  If it aids the rotation of the jump (as it should) it is very good, if it does not aid the rotation or inhibits it, is not very good.

Landing position is more complex.  It is very easy for a jump with very good takeoff and air position to have a poor landing position, even for clean fully rotated jumps.  This arises most often, in our observation, from an inability to control the body position as the rotation rate in the air must be reduced to zero on the landing and flow out of the jump.  It can also result from a lack of flow out of the landing (by landing on the toe, for example) resulting in breaks at the waste and other distortion of the body position.

From the there-or-not perspective, there seems no value in trying to assign a numerical metric to takeoff and landing body positions to quantitatively compare positions.

6) Element matches the music

This is the subjective opinion of the viewer, and not a biomechanical concept.