The 3D Serve:
Upward Swing Part 2
Brian Gordon, PhD
Our first article on the upward swing in the serve dealt primarily with the role of the torso or trunk. We identified the "cartwheel" action as most responsible for the transfer of forward angular momentum to the hitting arm.
We then discussed ways to alter the rotational speed of the trunk by changing its orientation in time and space. Finally we saw how the possible variations in the use of the trunk affected the positioning of the hitting arm at contact. (Click Here.)
We concluded that there are different ways to take the racket to the contact point, and these differences are a function of a series of specific joint rotations that can occur in different combinations.
But measuring and describing these components of the upward swing is different from explaining how the upward swing is actually executed.
Joint forces and torques are the engines that transfer momentum upward to the racquet. They drive the joint and segment rotations. But how do players make these joint and segment rotations happen? Let's address that in this article.
Let's start with a basic distinction and see what parts of the motion are driven by active muscle contraction and what happens as a consequence of other motions, what we've called dependent motion effects.
Finally, let's address some of the most common problems we see with the upward swing in coaching. We'll see how our 3D analysis allows us to pinpoint the problems and make subtle changes in the motion that can have a significant impact on racket speed.
The Mystery
The movement of the racket to contact is the subject of much conjecture and misunderstanding in the tennis world. It's not hard to understand why. Just look at the many elements involved: upper arm rotation, elbow extension, forearm supination and pronation, ulnar deviation, and wrist flexion.
We are talking about a very complicated motion. It is also a motion that happens very fast. During the upward swing the racket head speed can increase by up to 70mph in 1/10th of second. This is far too fast for the naked eye to observe accurately. The motion is so complicated that the full picture has yet to be fully described. Still, I believe we can identify pieces of the puzzle that will help any player.
Active Muscle Contraction
The first step in the process is to understand the causes of joint and segment rotation, what I call the "engines" of angular momentum transfer. The first is active muscle contraction. The second is what we have called the motion dependent effect, or motion dependent torque.
What do I mean by active contraction? When a muscle is contracted, it pulls on the adjacent bones. These bones are connected by a joint. The muscle contraction creates torque, called a joint torque, which causes the joint to either open or close.
The animation shows how this mechanism works. The muscle in the animation connects the forearm and upper arm. The arrows show what happens when the muscle is shortened by a contraction. The result is the rotation of one or both segments around the spanned joint. Put in simple English, in this case, the elbow extends.
Motion Dependent Effects
The second engine is what I call the motion dependent effect. This is a torque that is not created directly by muscular contraction. Rather it is created by joint torques and joint forces elsewhere in the chain of the motion.
We introduced this concept when we discussed racquet drop in the back swing. The animation shows how this works. In this hypothetical scenario, a muscular joint torque (the circular yellow arrow) is used to rotate the upper arm at the shoulder joint.
This rotation causes the upper arm to exert a force on the forearm at the elbow joint (the red arrow). If the force is not directed through the center of mass of the forearm (i.e. not aligned with the forearm bone(s), the forearm will tend to rotate. The result of the application of this force in the animation is that the elbow extends (the circular white arrow).
Making sense of these motion dependent effects is difficult. Our animation example here is actually over simplified, because there is also a joint force acting on the forearm at the wrist which plays a role. Another problem is that we don't as yet fully understand the origins of all the joint forces.
Nonetheless without fully understanding how they are generated, I believe it is still possible to examine the role of these joint forces in the creation of the upward swing. My hope is this explanation well shed some light on serve mechanics and answer some lingering questions that all players and coaches ponder.
Hitting Arm Kinetics
It is critical that players and coaches have a basic understanding of hitting arm kinetics when addressing stroke technique. This is because every player will use his own combination of muscular contraction (joint torque) and joint forces (motion dependent torque) to rotate the segments of the hitting arm.
To aid in the understanding of the kinetics of the hitting arm segment rotations, I've constructed an animation that allows the reader to investigate the causes of the rotations. The animation uses the data from my database of top college players.
For each hitting arm segment, you can select a rotation axis and a view. The rotation axis is displayed as a red arrow originating at the segment center of mass. The arrow corresponds to the joint axis of rotation (for example, the flexion/extension of the elbow). Another red arrow rotates around the axis. This rotating arrow simply describes a rotational direction around the axis, a direction I'll call positive.
As discussed, kinetic sources of a segment rotation may include muscular contractions around the joints at either end of the segment (proximal and/or distal). An additional source can be the turning effect of joint forces, again at either end of the segment.
The bars at the top of the animation indicate whether these kinetic sources tend to cause rotation in the direction of the red rotating arrow (positive), or in the opposite direction (negative). A green bar extending to the right indicates that the kinetic source is contributing to the positive rotational direction, shown by the rotating red arrow. A red bar extending to the left indicates that this source is contributing in the opposite direction.
The rotational effect of the joint forces can be hard to visualize. Therefore, the actual force at each joint is shown in the animation where applicable. Its rotational effect can be visualized by thinking of the arrow as a rope. Pulling the rope in the direction of the arrow can cause the segment to rotate around its center and this is the turning effect of that force.
This animation gives a real indication of how complex the kinetics during the upward swing really are. Many individual components interact to cause the motion we see. To make sense of it all, let's go segment by segment and see how all these factors work together, starting with the upper arm.
Upper Arm Non-Twisting Rotation
Upper arm rotation at the shoulder joint is driven primarily by conscious muscular torque throughout the upward swing. Early in the upward swing the elbow is both lifted (abduction: Y axis) and moved forward (horizontal adduction: X axis). In both cases, the motion dependent effect acts in opposition to the muscle torque.
Doesn't this mean the player is actually working against himself? Actually, this phenomenon is beneficial to muscular force production. The motion dependent torque allows the conscious contraction to occur in slower conditions than it would otherwise. As pointed out earlier in this series, muscle can produce more force if contraction is slower.
This is one of the scenarios often referred to by coaches as "loading" -- a muscle is pre-contracted--or opposed in contraction--due to effects of other body motions. In the case of the shoulder, the loading is a result of the joint force at the shoulder resulting primarily from cartwheel and twisting rotation of the trunk.
Upper Arm Twisting Rotation
Early in the upward swing, the shoulder externally rotates even though the internal rotator muscles are active. This is a continuation of the pattern discussed at length in the backswing. (Click Here.)
This pattern can be beneficial to muscle force production. Once internal rotation is initiated, it activates components of the stretch-shorten cycle. This internal rotation of the joint starts approximately half way through the upward swing. Active muscle contraction drives the shoulder internal rotation the rest of the way to contact.
Given the link between shoulder internal rotation and racquet speed around contact, maximizing internal muscle torque seems important. With 3D measurement we can verify the attributes of this change from external to internal rotation. But the efficacy of this transition can also be inferred by looking at the back view of a server.
In the interface watch the stick figure from the back view (center). Now step through the upward swing until the tip of the racquet reaches its furthest position to the right. Note the angle of the racquet shaft. Is it pointing directly downward or has it gone further? Higher level players generally rotate past vertical to the court. The racket shaft stays basically parallel to the tilt of their trunk. This indicates good external to internal transition. Notice that our junior player has less of a transition, with the racket tip pointing vertically directly down at the court.
Elbow Flexion/Extension
Rotation of the forearm at the elbow is important during the middle and late portions of the upward swing. Early in the upward swing the muscular torques are active, starting the extension of the elbow. There is also a contribution from the joint force at the elbow.
Past the midpoint of the upward swing, the contribution of the joint force at the elbow becomes much larger as the rate of extension increases. During this time, it is interesting to note that the muscular driven joint torque at the elbow changes to a flexion tendency.
It is possible that during this time the extensor muscles can't keep up with the rate of extension, creating an inhibiting influence on this action. A second possibility, though less likely, is that the flexors are used to control the rate of extension.
Either way, the mechanism links back to the actions of the trunk and shoulder. That is because they are the sources of the elbow force driving elbow extension. This is an attribute of high level performers, but is often lacking in developing players. In any case, as the racket approaches contact, the muscle driven elbow torque once again becomes the primary driver of elbow extension. Truly this is a complex chain of events, that, remember is happening within that 1/10th of second duration.
Pronation/Supination
In the popular literature, the term pronation is used to describe a vast array of motions. These include the turning of the racket after contact, and turning of the hand and wrist at various points of the swing. But anatomically, the terms pronation and supination specifically refer to the twist rotation of the forearm independent of the upper arm rotation.
Forearm pronation is controlled by joint torque throughout the upward swing. Initially in the upward swing, the forearm tends to supinate. However this effect is slowed by a pronator torque. This torque is responsible for the timing of the forearm pronation, such that it begins closer to contact. The slowing of supination through the pronator torque could represent another application of the stretch-shorten cycle.
Although pronation is a term that gets a lot of attention in coaching, its impact is arguably much less than the upper arm in creating racquet speed on the serve. The fact is that much of the forearm rotation observed by coaches is due to shoulder internal rotation, and not independent forearm pronation. The reality is that independent forearm pronation is used more for positioning the wrist joint and racquet head than for generating racquet head speed.
The Hand/Racquet
The hand and racquet are considered as a single segment in this discussion. They rotate as a function of wrist joint motion. The motion of the wrist during the upward swing is (surprise) also complex. You may recall that wrist extension was a significant contributor to racquet speed entering the upward swing.
During the mid portion of the upward swing, wrist ulnar deviation takes over as the most important wrist contributor to racquet speed. This is the motion of the hand in the direction of the pinky finger. This ulnar deviation is not caused by conscious contraction. It is driven primarily by joint forces and their motion dependent effects. These forces come primarily from the rapid extension of the elbow, as the ulnar deviation occurs in tandem with this extension.
Ulnar deviation has its greatest influence at the time elbow extension is being driven by the joint force at the elbow. This means that it is also driven by the actions of the trunk and shoulder that actually create the elbow joint force. This observation highlights how a seemingly simple joint motion can occur due to an intricate pattern of body motions -- motions that if not understood by coaches can yield sub-standard or even dangerous results.
Late in the upward swing, forearm pronation positions the wrist flexion axis so that wrist flexion can have the biggest impact on the forward component of racquet head velocity. And indeed, wrist flexion is a major contributor to racquet speed near and at contact. The significant portion of this flexion is caused by active and conscious muscular contraction and associated joint torque. Motion dependent effects have an inhibitory influence on this joint motion near contact.
Implications of Hitting Arm Kinetics
And there we have it: upper arm motion, external and internal rotation, elbow extension, ulnar deviation, pronation, and wrist flexion. Some parts driven by contraction, others not, forces that come from the hitting arm segment itself, and others from the previous links in the biomechanical chain-all in that magic 1/10th of a second.
So what does it all mean to players and coaches? I believe the lesson is that the tendency in coaching to focus on isolated pieces of the upward swing should be reevaluated. Statements such as "snap here" or "extend there" only make sense if made in the context of the overall motion and with an understanding of the unintended consequences.
For every action, there is a reaction. Advice to initiate a muscular contraction to move a joint in a certain way can actually do far more harm than good if other factors throughout the body have not been considered.
We have to keep this complexity in mind when we use great players as models for our strokes. A top player may have an isolated observable motion pattern that is predicated on many other factors that are not so readily observable.
What we need to understand is the big picture. The serve is built from the ground up. Angular momentum is generated from leg drive. This is optimized by body position, and transferred to the hitting arm by timely use of conscious muscular force. The rotational body sequencing produces motion dependent effects, which dictate further muscular contraction -- all this leads to contact racquet speed and direction. Again, it's a complex picture and it's a mistake to isolate one or two factors as the magic key to high performance serving.
Common Upward Swing Problems
We've covered a lot of issues in part 1 and part 2 of the upward swing. Now let's apply what we've learned to some common upward swing problems I see in coaching, and pose some possible solutions. In this way perhaps we can see how technical analysis can actually come into play in coaching practice.
Despite the vast array of things that could go wrong, amazingly, the common problems that actually occur are within a fairly narrow range. And fortunately, we now have a clear way to gauge when problems occur and in what specific part of the upward swing.
This is because, in our 3D data, the errors will nearly always manifest themselves as irregularities, or plateaus, in the curve charting racquet speed. So if the curve doesn't show smooth continuous acceleration, we are suspicious the player may not have maximized his racket speed.
These plateaus can occur as a result of insufficient, mistimed, or nonexistent joint rotations in the kinetic sequencing of events. Care must be taken to identify these errors early in development because they are much more difficult to correct later. Worse, in many cases they can also expose players to potential injury.
Insufficient Upper Trunk Twist Rotation
Insufficient upper trunk twist is the most common error that can reduce racquet head speed development by 10% or more. Further, it diminishes the force placed on the upper arm at the shoulder joint, a force that, as we have repeatedly seen, is important to the sequencing of the various kinetic events.
The biggest related problem is that if there is insufficient trunk twist, this racquet speed source must be replaced. It turns out that the way players do is this is by adding independent (non-twist) rotation of the shoulder joint. Repetitive and excessive use of the shoulder joint in this fashion gives us concern for injury.
The causes of the insufficient upper trunk twist rotation can be traced to problems in generating or transferring angular momentum during the backswing. These problems generally relate to either timing problems in the leg drive, or in the case of many junior players, insufficient power in the required muscles. (This is why off court training is so important.)
Hitting Arm Motion Sequencing
Let's review the chain of events for the hitting arm. First shoulder joint motion to elevate and move forward the elbow joint. Next there is elbow extension along with ulnar deviation. Finally there is internal shoulder rotation along with wrist flexion. Deficiencies along this chain have a negative impact on racquet speed development, which we can see in the velocity curves.
The most common breakdown in this chain at the beginning, the initial shoulder joint motion. Due to lack of strength or poor technique, the elbow is never correctly positioned during the upward swing. This means it does not move upward and/or forward from the shoulder joint.
This can happen for two reasons. First, the elbow positioning occurs strictly as a function of trunk rotation without independent shoulder joint motion. Second, the player substitutes early elbow extension for the independent shoulder joint movement.
In the first case the result is that the orientation and direction of the elbow extension reduces the contribution to racquet speed. In the second case, the early elbow extension throws off sequence of racket speed development. In effect, the elbow extension is contributing at the wrong time. This creates a plateau in velocity in the later stages. The effect of the elbow extension has been depleted at the time it should be playing it's critical role. This leaves a void in the middle portion of the upward swing.
The latter deficiency is often a result of an unconscious effort by the neuromuscular system to bypass a weak link--to compensate for the absence of shoulder joint motion by substituting the next link in the chain. But the effect here is always the same, the players ends up supplementing the late racket speed development with extra non-twisting shoulder joint motion.
Shoulder Internal Rotation
My experience has been that the ability to integrate explosive internal shoulder rotation into the upward swing is the x-factor in developing a high level serve. It has also been my experience poor utilization of this resource dooms developing players to mediocrity.
To a large degree, developing a player's use of this action relates to natural ability, although this can be enhanced significantly by a well designed off-court training program. This is not the place to go into the details of this type of training, although I plan to write more about that later. But there is no doubt that regardless of the natural ability to internally rotate the shoulder, the mechanics themselves play an important role.
To review, the external rotation in the backswing, (which also continues into the early upward swing) allows the use of a stretch-shorten cycle. This occurs when the internal rotating muscles are used to slow the external rotation. This means the internal rotators are activated and put on stretch before the actual internal rotation of the segment begins later in the upward swing. Recall that the effectiveness of this mechanism can be estimated by viewing the angle of the racquet in a back view. We want to see the racket shaft in line with the angle of the torso.
Racket to Arm Angle
The configuration of the body segments near contact is also critical. No amount of internal shoulder rotation will move the racquet forward unless the proper racquet to arm angle is used. This angle is also related to other angles in constructing the contact body configuration.
This has been one of the main areas of emphasis in my work with the junior shown in the user interface. To see the effect of improved positions, select "Compare Trials" and the date 111106 in the left margin of the interface.
Specifically we have worked with our model player to decrease lateral trunk tilt, and that has met with some success. In the comparison mode select "Display Angle View" -- then use "Show Next Angle" to toggle through to the contact angles. It shows how the trunk is less tilted in the more recent serve (49 vs. 44 degrees from the left horizontal).
Let's see how that relates to the critical hitting arm angle. Toggle two more angles and notice the angle between the hitting arm and racquet has decreased from 165 to 148 degrees. The result is that the contribution of internal rotation to racquet head speed is increased.
Select "Velocity Graphs" from the "Graphical Displays" menu, then select "Upper Arm Twist Rotation" in the selections below the graph. The numbers show that the contribution to racquet speed from internal rotation increased from 3 to 8 MPH. Still, 8 MPH is only 13% of the 62 MPH racquet speed so more work is needed.
The improved angle of the hitting arm to the racket has another benefit, because it can also reduce the contribution of the non-twisting shoulder contribution over the same interval. This can be verified by selecting "Shoulder Joint Motion" also below the graph.
The Follow Through
The foremost goal of the follow through is to gracefully slow the segments so that injury can be avoided. To accomplish this, the range of slowing should be made as long as possible. Because slowing the segments often requires contracting muscles that are lengthening (eccentric contraction), the high forces involved should be spread over the longest possible duration.
Of particular interest here is the shoulder internal rotation. Post-contact, the external rotator muscles will be eccentrically activated to slow the internal rotation. Because the body will attempt to extend the duration of this slowing for safety, the position attained can be readily observed and used as a rough indicator of the extent to which internal rotation was used leading into contact. The hitting side of the racquet face pointing at least towards the right sideline indicates strong internal rotation.
The dynamic elbow extension advocated in this article comes with certain risks. At the point of full extension, there can be extensive bone on bone contact between the forearm and upper arm. The risk of joint damage can be reduced by a technique seen in many players, but lacking in others.
Research done in pitching discovered a quick flexion of the elbow following the rapid elbow extension into release. This flexion combined with shoulder internal rotation already present, decreases the magnitude of the bone on bone contact at full extension. My experience is that is something that often needs to be trained. You can see this very readily in the Sampras motion where the elbow bends almost immediately after contact and earlier than virtually any other player.
Recovery from the serve involves quickly regaining balance and control of the body. This goal is in stark contrast of the mechanics to create racquet speed which imposes rotational chaos to the body. The amount of angular momentum that is required to generate racquet speed will create a less than graceful transition into the point if not dealt with in the recovery.
In the same way angular momentum was transferred into the trunk during the hitting phases, it can be transferred out during the recovery phase. This is accomplished by redistributing the body momentum back to the legs, and specifically the trailing leg.
This is the reason for the leg kick back we see all the top players used. The kick back transfers angular momentum to this leg causing the rest of the body to rotate more slowly, and allowing smoother recovery.
The extent and direction of the leg kick also speaks volumes about how angular momentum was generated earlier. It's a good key for diagnosing what actually happened in the motion. The lack of kick back indicates insufficient angular momentum to hit a big serve. A kick back to the side indicates too much lateral angular momentum, something that should be addressed immediately. The preferred motion is a strong, straight back kick. This indicates ample forward angular momentum.
Parting Shots
My goal is initiating this series of articles was to begin bridging the gap between the research concepts in tennis sport science publications and the real world of player development. I tried to do this in two ways. First, by trying to explicate some findings from my research and those of others. Second, by showing how this work can be integrated in the coaching work we do at our training facilities.
In this process I found myself walking a fine line between strict mechanical precision and making the work comprehensible to players and coaches with a wide range of backgrounds. To some extent this may have been a zero-sum game. Some readers probably found it lacking adequate technical specificity, while many others found it still quite complicated. Nonetheless I believe we have succeeded in highlighting the many factors and choices that go into constructing a high level serve for yourself or for your players.
One Note of Caution
All these factors require dynamic patterns of motions. This brings me to a final word of caution, which is that mastering them requires preparation of the body in terms of strength, conditioning, and flexibility. If something feels as if it is causing you, or might cause you an injury, it probably will. Investigate this for yourself with your coach or a strength trainer and use your instincts and common sense before making radical changes in your motion in your desire to serve like a top pro.
