Measuring racket head speed: the Sampras serve

[gzhpcu]

John,
Very interesting article! Lots of information.

The high acceleration of the racket prior to contact:

Isn't this basically because of the following?

The arm itself, prior to impact, slows down, moves very little, but the racket swivels at the hand (this is why a loose wrist is essential)... The arm is slowing down, creating a whip-like motion of the racket, and the path of the racket is up and across the ball. I think this is what Bungalo Bill once brought up in an article here (except that I can't find it...).

[johnyandell]

Here is the Bungalow Bill article:

http://www.tennisplayer.net/members/tour_strokes/bungalow_bill/waves_and%20_the%20_myth_wrist/waves_and%20_the%20_myth_wrist.html

As for the arm--really doubt it slows down--remember you have 1/10th of a second! Also it can't swivel at the wrist or very little--the forearm is what rotates it. The wrist is flexing forward from the laid
back position at the drop--and the upper arm and forearm are rotating like crazy.
That's driving the flex of the wrist which should stay loose--but probably like the forehand that movement of the wrist is actually inhibited by the player somewhat to set the exact racket angle.

In any case as with all these strokes the dark mysteries of the how and why are not fully understood. The key is to make the key positions with the right checkpoints. And what players tell themselves when they do that doesn't necessarily have to correspond to reality. This is why you hear so much crazy talk--or one reason anyway...

[gzhpcu]

What do you think of this John?

https://www.youtube.com/watch?v=nKhbKSKKmyM

[kenh]

Sampras serve

In going forward on this, to fully explain the sudden and strong force into the ball, it might be interesting to correlate the timing of the full contraction of the stomach muscles, which are transferring the forces from pushing off the ground, with the timing of the pronation into the ball. The core relaxation could also be part of explaining the deceleration.
As a side thought, it might be interesting to compare the racquet acceleration to the acceleration of a pro boxing punch or karate punch. Both can have what may be similarly strong acceleration and deceleration over a short period of time.

[gzhpcu]

Not to forget the leverage of the rotating racket up and across...

[johnyandell]

Phil,
He's got the position right but it's not the forearm it's the upper arm.

KenH,
Interesting questions. No data on any of them that I am aware of. My view is that if the technical positions are good, most other things take care of themselves.

[curiosity]

It seems to me like this, as a partial reflection on Pete's serve: The serve (as, too, the forehand) at some point employs some generation of linear momentum in order to gear up the RHS that can be generated during the crucial angular momentum-increasing actions. The rotary angular-momentum phases/segments can achieve much greater acceleration on their own than can pure linear moves (though the linear moves are often impure, mixed with some rotary/angular motion). Together and properly sequenced "mainly linear" and "mainly rotary/angular" motions are the complex sources of racquet head velocity, it seems safe to say.

The enemies, then, of a continuously accelerating (positive second derivative-possessed service motion, one producing a racquet swing that constantly wants to pull out of your hand and fly skyward), are well known. Hitches, moments of energy dissipation during which deceleration occurs, swing paths which take uselessly odd paths around and up to the ball, clumsy final-instant shoulder/arm/wrist/hand actions which commit several of the previously listed sins...all these enemies need to be eliminated.

Back to Pete: It has long struck me that Pete's circular loop back and up induced no or little waste from needless direction changes. His very much "start slow to finish fast" pacing seemed so well calculated to set up the greatest part of the continuous 2nd derivative's acceleration within the last instants of his motion up to contact, with no breaks in acceleration along the way.

This is really just a restatement in other terms of the points JY made in his "amazingly, 65% of the RHS is produced in the last .5 second,"

Obviously, even though the majority of speed is produced in the last .5 second, that does not mean we can produce that speed without performing the correct actions at the correct pace in the prior 1.5 seconds. Much set up and enforced timing is mandatory. This is equally true in the forehand. In both serve and forehand if we start too fast we'll hit at least one dissipation point. Starting slow isn't an option in most forms of service, but is a requirement.

Or so it seems to me.

[gzhpcu]

The serve is the most complex stroke in tennis, and, looking at many different pro servers, you often get the impression that there is more than one way to skin a cat. Consider the following:

Roscoe Tanner

Pancho Gonzalez

Pete Sampras

Roger Federer

Stefan Edberg

Taylor Dent

Milos Raonic

Ivan Ivanisevic

So the puzzle is: the whole kinetic chain is complex, and some players accentuate some aspects more than others.

What are the core commonalities? What aspects are coupled? (Accentuating one aspect to the detriment of another?) Is a perfect model possible? How much is tied into the player's biomechanical physical makeup and potential?

[johnyandell]

Good questions and I have no short answers. But all those guys share something like that 1/10th of a second racket swing path.

[curiosity]

It occurs to me that the sudden explosion of speed in the last 1/10th of a second, propelling the racket head to 90 MPH, could not be otherwise. That is to say, if 90 MPH were reached earlier in the swing it could never be maintained.

It seems that most of the service motion is designed to coordinate the toss and the swing, and then to generate racket head speed, spin, and direction. Since speed, spin, and direction can only be imparted at contact, that is where all designs to produce speed and spin must peak.

I would guess the statistic for the forehand is somewhat similar, with a great percentage of the RH speed produced in the last instants, and for the same reason: The speed must be produced as close as possible in time to the instant of contact with the ball, else the speed will dissipate before contact.

[johnyandell]

Curiosity,
Interesting thoughts as usual. And you are most likely right about the forehand. I say most likely because that type of analysis amazingly does not exist. Working with a group now with some new technologies and hope to have similar data on the forehand and other strokes this year and following--maybe not of Pete or players at that level, but a world view.

[curiosity]

JY: Good. I look forward to such data.

[gzhpcu]

Some observations:

Looking at the animation, you can see that the body is a complex link system, each part contributing to the speed of the tip of the racket, some more, some less.

The maximum contribution comes from the arm link system: shoulder - elbow - wrist - racket. The transfer of speed goes down the chain, each component braking to augment the acceleration of the next component, finishing ultimately with the racket head. Seems pretty much like a whipping action. The arm needs to be very loose.

The rest of the body motion does not seem to contribute much to the acceleration. Hips move forward, legs forward and up. Seems like their contribution is mostly to increase the mass behind the racket at impact.

[stotty]

Good post...

Some observations:

Looking at the animation, you can see that the body is a complex link system, each part contributing to the speed of the tip of the racket, some more, some less.

The maximum contribution comes from the arm link system: shoulder - elbow - wrist - racket. The transfer of speed goes down the chain, each component braking to augment the acceleration of the next component, finishing ultimately with the racket head. Seems pretty much like a whipping action. The arm needs to be very loose.

The rest of the body motion does not seem to contribute much to the acceleration. Hips move forward, legs forward and up. Seems like their contribution is mostly to increase the mass behind the racket at impact.

This is a good post. A great summary of events. It's interesting that a serve deficient in certain areas can impact the force of that whipping action you speak of - insufficient trunk rotation will reduce the power from the upper arm and shoulder and reduce racket speed, for example.

It seems to me right from the very outset each part of the serve influences the next part.

[klacr]

This is a good post. A great summary of events. It's interesting that a serve deficient in certain areas can impact the force of that whipping action you speak of - insufficient trunk rotation will reduce the power from the upper arm and shoulder and reduce racket speed, for example.

It seems to me right from the very outset each part of the serve influences the next part.

Only as strong as its weakest link.

Kyle LaCroix USPTA
Boca Raton