A New Teaching System: The Serve: Role of the Legs

[stroke]

I would be interested in seeing (with good players) end point racquet head speed comparisons with and without leg drive. I'm not sure they would be drastically different.

I've seen way too many serves north of 100 miles an hour, without getting up off the ground, to believe that it is an absolute requirement to access speed. Having said that, why would a player not want to recruit the bigger muscles groups, push harder against the ground (for more available ground reaction force), and achieve a higher contact point? No, IMO the role of the legs is about better angles (higher contact point) injury prevention, and efficiency in creating speed. I couldn't imagine what it would be like serving the last 40 years without working against the ground, in a significant way.

But here's the thing. There is no better way to screw up a serve (hitches, power outages) than to start to bring the legs into play (in any significant way) when they are out of sync with the arm progressions. It absolutely destroys a serve, so special attention is needed to that relationship.

Interestingly, Pat Dougherty, the Serve Doc, references this same thought in teaching what he calls "the serve doc's preferred method" of serving in his MPH instructional serve DVD. I think it is the best serve instructional video I have seen.

[johnyandell]

The 3D studies Bruce Elliot of pro players did showed much more stress on the arm and shoulders for players with less knee bend.

My opinion is mph is misleading. It takes racket head speed for spin as well. The legs are key is producing the combo for a heavy serve as compared to a fast one. And save the upper body joints.

[tennis_chiro]

Get your thinking caps on!

Okay, boys and girls, you are going to have to put on your thinking caps to follow me on this one, but I hope you will. I really want to see some blowback on what I have to say here. I don't mind agreement at all, but I expect I'm going to get something else.

I'm pretty much in line with what 10splayer has to say about too much emphasis on the leg drive screwing up a good service action before it ever gets developed. Furthermore, I am convinced the primary power from the lower body is the result of rotation of the hips rather than from the extension of the knees. Certainly, leg drive and knee extension is part of it, but a much smaller part of the speed than the emphasis placed on developing leg drive would have you believe.

First of all, examine the slow motion video of any of the top servers. By the time they contact the ball and usually by the time the arm makes it's final explosive move upward in that last tenth of a second when most of the acceleration of the racket head takes place, the player is no longer rising; often they are already coming down, but at least they are no longer rising. Check the top of the head against the background in a side view. So the upward propulsion of the legs didn't actually add any direct velocity to the upward speed of the racket head. But that leg power had to contribute something. How did that power get transferred to the racket?

The power of the legs is stored through the kinetic chain in the shoulder/arm/upper torso-hitting joint. Essentially, all the preparation pulls the slingshot back and then at the last minute we release it. Without the leg drive, you can't pull the slingshot back as far. By the same token, if you place your arm in the "power position" (I don't know if I have the right term, but I mean the position where the racket head reaches its lowest point in the loop behind the back before going up to the ball; maybe the "pro drop position" would be better), you can't get it as far back as you can when you drop it into that position from the "trophy position" and use the weight and momentum of the racket head to pull the muscles of the shoulder/arm/etc a little further back (the slingshot). This is on top of the small amount of racket head speed that comes from the racket head already moving when it gets to this point. Just put your racket in that "pro drop position". You can only go so far, but someone can then pull on your racket head and stretch your muscles a little further. That extra stretch is what the leg drive is accomplishing.

But there is a law of diminishing returns here. Just as your friend can only pull your racket a little lower than the position you can place it in yourself, that friend can only go so far before limitations of the joints and soft tissue keep you from going any farther without injuring you. So when you have a small bowing action like Michael Stich, you accomplish probably 90% of the shoulder stretch that you do with the violent greater leg drive that occurs in an action like Roddick (that's a totally subjective assessment just from me). Certainly, you can jump higher with the deeper knee bend, but I'm not so sure the initial acceleration of the deep knee action is as quick as the bounce that Stich can get by simply "bowing" the legs just a bit. The amount you stretch the slingshot is going to be a function of how quickly we accelerate that racket with the leg action at the bottom of the "pro drop". It's only a couple of inches at most; maybe with the rotation of the arm in the shoulder joint, it could amount to a couple more than that. Each extra degree of external rotation prior to your internal rotation is like one mph. Understand that the arm is leaving the racket head behind as the legs extend up; I think initial acceleration (of the leg drive or bowing or whatever) is more important than the final speed of the jump that is achieved.

So while I might be able to double the distance I can explode off the ground with a good leg drive from a platform stance, I am only going to marginally increase the actual additional pull I get on my "slingshot". I will have a higher contact point and that has real value. But I am also making the whole action of meeting the ball a lot more difficult to coordinate and repeat consistently enough to give me the consistency and accuracy that is an essential requirement of a good serve.

I don't usually worry about whether my students adopt a platform or pinpoint stance, but I really like the motions of people like Gonzales, Hoad or, more recently, Stich and Krajicek. There's a whole different discussion we have to get into about the best way to develop a consistent repeating toss that's synchronized with the rhythm of the weight transfer/rock, but while related, it is an entirely different discussion.

But I've been trying to come up with a rationalization of why I prefer the pinpoint or getting all the weight to the left side earlier than you do with the centered platform approach. I think it comes down to pulleys and levers and fulcrums. We can at least agree that the body generates force exclusively through pulleys and levers and fulcrums. I think we hit the serve against the fulcrum of the left side (right handers). So when I push up the left side in a more pinpoint configuration, the force is going more through my left hip and left shoulder. It seems to me that this would contribute to helping me pull the slingshot a little further back from the fulcrum of the left side. So the right clavicle is stretched a little further away from the sternum, etc. If the upward force goes through the middle of my body as opposed to up the left side, then there is no stretch of the slingshot of the right side away from the left side. I want to pull as many of those components of that slingshot as I can while maintaining a very calm left side that can accomplish the tossing of the ball consistently; then I want to let that slingshot fire its payload at the last possible minute.

Bear with me a moment. Imagine you are pulling an arrow in a bow with your right hand; it won't amount to anything if your left hand can't hold the bow steady and provide something to pull against.

To me it is critical that you develop the ability to hit solidly "against the left side" in the manner I am describing before you start compounding the motion with leg drives that destabilize the whole system. The USTA.TV clips for quickstart/ROG learning landmarks (check out the usta.tv app; it's pretty cool) lay out a wonderful progression that shows how a kid can and should develop pretty big leg drive right from the beginning. It looks great on the videos. But I'm not at all convinced that is the right way to develop a good service motion.

As 10splayer said quite eloquently in my opinion,
"There is no better way to screw up a serve (hitches, power outages) than to start to bring the legs into play (in any significant way) when they are out of sync with the arm progressions. It absolutely destroys a serve, so special attention is needed to that relationship."

Stotty has had a bunch of posts on this thread that I would pretty much agree with. Clearly, the kick back has nothing to do with generating power. It is a balancing mechanism that takes place after the power is generated and used. It may be necessary because we can't go to the net that often, but I seriously question any proposition that Gonzales or Hoad would have served better with a kickback of the right leg. And frankly, if I could get someone to serve like Hoad or Gonzales, I don't think I would mess with their motion at all. I'd release them to leave the ground when they feel the need and try to get them to drive up perhaps a little more; but if they were still on the ground, I'd probably leave them be.

Final Argument about Kickback:
There is one single human embodiment of natural, efficient, powerful free motion on a tennis court that stands out for me above all others, and although many of his contrived motions may have been distasteful, Ilie Nastase did not have a kickback or a deep knee bend to his serve!

don

[stotty]

Great post, tennis_chiro.

Nastase reached up but didn't "drive" up with the same force as players do today...as can be seen on the very first point of this clip.

http://www.youtube.com/watch?v=lf4wrrpzdYc

I downloaded the clip on to my desktop and toggled through it frame by frame. Players like Nastase and Stich seem fall through their serves rather than driving upwards as much as some players. I guess if you have a knee bend deeper than a certain point you cross a threshold where you have to drive up significantly more to come out of it...resulting unavoidable kickback.

When you look at Nasty serving he looks like a cross between two eras. One imagines he spent his formative years under the law of having to keep his left foot on the ground, and his remaining years trying to get used to the freedom of being able to leave the ground. He barely leaves the ground at all in the clip.

[tennis_chiro]

Was really hoping for more responses on this post!

Okay, boys and girls, you are going to have to put on your thinking caps to follow me on this one, but I hope you will. I really want to see some blowback on what I have to say here. I don't mind agreement at all, but I expect I'm going to get something else.

I'm pretty much in line with what 10splayer has to say about too much emphasis on the leg drive screwing up a good service action before it ever gets developed. Furthermore, I am convinced the primary power from the lower body is the result of rotation of the hips rather than from the extension of the knees. Certainly, leg drive and knee extension is part of it, but a much smaller part of the speed than the emphasis placed on developing leg drive would have you believe.

First of all, examine the slow motion video of any of the top servers. By the time they contact the ball and usually by the time the arm makes it's final explosive move upward in that last tenth of a second when most of the acceleration of the racket head takes place, the player is no longer rising; often they are already coming down, but at least they are no longer rising. Check the top of the head against the background in a side view. So the upward propulsion of the legs didn't actually add any direct velocity to the upward speed of the racket head. But that leg power had to contribute something. How did that power get transferred to the racket?

The power of the legs is stored through the kinetic chain in the shoulder/arm/upper torso-hitting joint. Essentially, all the preparation pulls the slingshot back and then at the last minute we release it. Without the leg drive, you can't pull the slingshot back as far. By the same token, if you place your arm in the "power position" (I don't know if I have the right term, but I mean the position where the racket head reaches its lowest point in the loop behind the back before going up to the ball; maybe the "pro drop position" would be better), you can't get it as far back as you can when you drop it into that position from the "trophy position" and use the weight and momentum of the racket head to pull the muscles of the shoulder/arm/etc a little further back (the slingshot). This is on top of the small amount of racket head speed that comes from the racket head already moving when it gets to this point. Just put your racket in that "pro drop position". You can only go so far, but someone can then pull on your racket head and stretch your muscles a little further. That extra stretch is what the leg drive is accomplishing.

But there is a law of diminishing returns here. Just as your friend can only pull your racket a little lower than the position you can place it in yourself, that friend can only go so far before limitations of the joints and soft tissue keep you from going any farther without injuring you. So when you have a small bowing action like Michael Stich, you accomplish probably 90% of the shoulder stretch that you do with the violent greater leg drive that occurs in an action like Roddick (that's a totally subjective assessment just from me). Certainly, you can jump higher with the deeper knee bend, but I'm not so sure the initial acceleration of the deep knee action is as quick as the bounce that Stich can get by simply "bowing" the legs just a bit. The amount you stretch the slingshot is going to be a function of how quickly we accelerate that racket with the leg action at the bottom of the "pro drop". It's only a couple of inches at most; maybe with the rotation of the arm in the shoulder joint, it could amount to a couple more than that. Each extra degree of external rotation prior to your internal rotation is like one mph. Understand that the arm is leaving the racket head behind as the legs extend up; I think initial acceleration (of the leg drive or bowing or whatever) is more important than the final speed of the jump that is achieved.

So while I might be able to double the distance I can explode off the ground with a good leg drive from a platform stance, I am only going to marginally increase the actual additional pull I get on my "slingshot". I will have a higher contact point and that has real value. But I am also making the whole action of meeting the ball a lot more difficult to coordinate and repeat consistently enough to give me the consistency and accuracy that is an essential requirement of a good serve.

I don't usually worry about whether my students adopt a platform or pinpoint stance, but I really like the motions of people like Gonzales, Hoad or, more recently, Stich and Krajicek. There's a whole different discussion we have to get into about the best way to develop a consistent repeating toss that's synchronized with the rhythm of the weight transfer/rock, but while related, it is an entirely different discussion.

But I've been trying to come up with a rationalization of why I prefer the pinpoint or getting all the weight to the left side earlier than you do with the centered platform approach. I think it comes down to pulleys and levers and fulcrums. We can at least agree that the body generates force exclusively through pulleys and levers and fulcrums. I think we hit the serve against the fulcrum of the left side (right handers). So when I push up the left side in a more pinpoint configuration, the force is going more through my left hip and left shoulder. It seems to me that this would contribute to helping me pull the slingshot a little further back from the fulcrum of the left side. So the right clavicle is stretched a little further away from the sternum, etc. If the upward force goes through the middle of my body as opposed to up the left side, then there is no stretch of the slingshot of the right side away from the left side. I want to pull as many of those components of that slingshot as I can while maintaining a very calm left side that can accomplish the tossing of the ball consistently; then I want to let that slingshot fire its payload at the last possible minute.

Bear with me a moment. Imagine you are pulling an arrow in a bow with your right hand; it won't amount to anything if your left hand can't hold the bow steady and provide something to pull against.

To me it is critical that you develop the ability to hit solidly "against the left side" in the manner I am describing before you start compounding the motion with leg drives that destabilize the whole system. The USTA.TV clips for quickstart/ROG learning landmarks (check out the usta.tv app; it's pretty cool) lay out a wonderful progression that shows how a kid can and should develop pretty big leg drive right from the beginning. It looks great on the videos. But I'm not at all convinced that is the right way to develop a good service motion.

As 10splayer said quite eloquently in my opinion,
"There is no better way to screw up a serve (hitches, power outages) than to start to bring the legs into play (in any significant way) when they are out of sync with the arm progressions. It absolutely destroys a serve, so special attention is needed to that relationship."

Stotty has had a bunch of posts on this thread that I would pretty much agree with. Clearly, the kick back has nothing to do with generating power. It is a balancing mechanism that takes place after the power is generated and used. It may be necessary because we can't go to the net that often, but I seriously question any proposition that Gonzales or Hoad would have served better with a kickback of the right leg. And frankly, if I could get someone to serve like Hoad or Gonzales, I don't think I would mess with their motion at all. I'd release them to leave the ground when they feel the need and try to get them to drive up perhaps a little more; but if they were still on the ground, I'd probably leave them be.

Final Argument about Kickback:
There is one single human embodiment of natural, efficient, powerful free motion on a tennis court that stands out for me above all others, and although many of his contrived motions may have been distasteful, Ilie Nastase did not have a kickback or a deep knee bend to his serve!

don

I really hoped I would get a little more response on this idea that the legs can only store so much energy in the shoulder efficiently and bending the knees more than a simple bow may get you higher, but it won't actually add any power and may detract from your consistency and accuracy.

Come on guys. Shoot me down! ... Or agree!

don

[bottle]

I agree, but my arthritic front leg may make my opinion suspect. I've had to diminish thrust from that leg and stop hopping on it, but I still can brace with it to put fulcrum on left side. I like that idea very much, and my serves overall have improved.

However, people tend to think whatever they presently are doing is just peachy.

I have one rear foot serve where I immediately kick the weak leg back with everything barely off the ground. That certainly puts rotational center at center of body rather than on left side. Not as solid but a quick surprise even though ball may sit up waist high.

Enough about me. Like you, I'd like to hear somebody else's opinion too.

[stotty]

I really hoped I would get a little more response on this idea that the legs can only store so much energy in the shoulder efficiently and bending the knees more than a simple bow may get you higher, but it won't actually add any power and may detract from your consistency and accuracy.

Come on guys. Shoot me down! ... Or agree!

don

I got lost here and there with your post. But I agree (I think) if you are saying bending the knees more will get you into positions that you cannot achieve with less bend.

I agree with the kinetic chain business and especially in how it all uncoils. This article video sums it up well:

http://www.tennisone.com/membership/coursepromo/kickserve/5.servesales.as.php

I find "saving the hit" interesting in the side by side comparison of Sampras and Murray.

I will add more to this post shortly....

[don_budge]

Wawrinka Videos?

The videos of the Stanislas Wawrinka serve are not working in my computer. But watching his delivery he doesn't appear to have any kind of exaggerated leg thrusting going on.

[tennis_chiro]

Sorry

I got lost here and there with your post. But I agree (I think) if you are saying bending the knees more will get you into positions that you cannot achieve with less bend.

I agree with the kinetic chain business and especially in how it all uncoils. This article video sums it up well:

http://www.tennisone.com/membership/coursepromo/kickserve/5.servesales.as.php

I find "saving the hit" interesting in the side by side comparison of Sampras and Murray.

I will add more to this post shortly....

Sorry it is so complicated, but I would need a whole animation company to properly illustrate what I am trying to postulate as an explanation here. Basically, I am saying that the extreme leg/knee bend creates a more complicated motion without a significant enough increase in power to justify the lowered consistency. Stotty, I hope you will try again to wade through my long post there.

As for Doug King's videos and service course, it looks really interesting. I got a couple of great ideas just looking at his "teaser". However, I disagree that the shoulder/hip turn is saved until the last instant. It's too slow. It has to happen before the faster last instant action of the arm. Perhaps it is still completing at impact, but even in the videos of Sampras and Murray that King uses, you can see that Pete has completed most of the hip/shoulder turn before the arm makes that last 90degrees plus move up to the ball. Pete is definitely not as much facing the net as other servers, but he starts much further back (facing the back fence). In Doug's analysis of why Murray's second serve is so much slower than his first, it would have been better to compare Murray's first serve to his second instead of to Pete's serve. But King's theory/paradigm about short lever power looks really interesting.

don

[stotty]

Sorry

Sorry it is so complicated, but I would need a whole animation company to properly illustrate what I am trying to postulate as an explanation here. Basically, I am saying that the extreme leg/knee bend creates a more complicated motion without a significant enough increase in power to justify the lowered consistency. Stotty, I hope you will try again to wade through my long post there.

As for Doug King's videos and service course, it looks really interesting. I got a couple of great ideas just looking at his "teaser". However, I disagree that the shoulder/hip turn is saved until the last instant. It's too slow. It has to happen before the faster last instant action of the arm. Perhaps it is still completing at impact, but even in the videos of Sampras and Murray that King uses, you can see that Pete has completed most of the hip/shoulder turn before the arm makes that last 90degrees plus move up to the ball. Pete is definitely not as much facing the net as other servers, but he starts much further back (facing the back fence). In Doug's analysis of why Murray's second serve is so much slower than his first, it would have been better to compare Murray's first serve to his second instead of to Pete's serve. But King's theory/paradigm about short lever power looks really interesting.

don

It's beyond me to truly fathom what body parts generate what quantitatively in a serve. I find the biomechanics of serving incredibly complex, and each individuals serve is unique. Like you say it's all levers, pulleys and fulcrums...difficult stuff. I read your post twice but honestly didn't feel qualified enough to offer an informed opinion.

I think Tennisplayer is great but it centers mostly around modern tennis. There are three types of stance: pin point, platform...and, well, "crossover", for want of a better word. Plenty of players "crossed over" in the past. Back in the days of "one foot on the ground" everyone did it. Stich and Nastase (and many others from the past) crossed their right foot over there left as they served volleyed their way to the net. They had no kick back as such...just less knee bend and more forward momentum instead. What's wrong with that? It's annoying when perfectly valid serves get deleted. I wonder if some of these servers weren't more consistent. I once saw Nastase serve four aces in a row to win a game. He served well!

I cross over and have no kick back. I have a friend who serves much better than me who crosses over and he doesn't have kick back either. We both serve and volley...on both first and second serves. Neither of us breaches the knee-bend threshold to force an upward drive that would create kick back by default.

I do think the deeper knee bend may be necessary for a bazooka serve. Stich was around the 130mph mark on his best first serves. The 140mph and upwards club are probably all deeper knee-benders than Stich or Nastase ever were.

Just my two cents...sorry I cannot offer more in the biomechanics department.

[tennis_chiro]

It's beyond me to truly fathom what body parts generate what quantitatively in a serve. I find the biomechanics of serving incredibly complex, and each individuals serve is unique. Like you say it's all levers, pulleys and fulcrums...difficult stuff. I read your post twice but honestly didn't feel qualified enough to offer an informed opinion.

I think Tennisplayer is great but it centers mostly around modern tennis. There are three types of stance: pin point, platform...and, well, "crossover", for want of a better word. Plenty of players "crossed over" in the past. Back in the days of "one foot on the ground" everyone did it. Stich and Nastase (and many others from the past) crossed their right foot over there left as they served volleyed their way to the net. They had no kick back as such...just less knee bend and more forward momentum instead. What's wrong with that? It's annoying when perfectly valid serves get deleted. I wonder if some of these servers weren't more consistent. I once saw Nastase serve four aces in a row to win a game. He served well!

I cross over and have no kick back. I have a friend who serves much better than me who crosses over and he doesn't have kick back either. We both serve and volley...on both first and second serves. Neither of us breaches the knee-bend threshold to force an upward drive that would create kick back by default.

I do think the deeper knee bend may be necessary for a bazooka serve. Stich was around the 130mph mark on his best first serves. The 140mph and upwards club are probably all deeper knee-benders than Stich or Nastase ever were.

Just my two cents...sorry I cannot offer more in the biomechanics department.

Thanks for the feedback, Stotty. We seem to be in basic agreement here. But I was hoping some of the more biomechanically eriudite posters here (Doug, Brian?) might take my supposition apart (or not) that there is only so much energy that can be stored in the "serve slingshot" and that there is a point of diminishing returns in creating more power with increased leg/knee action, especially when you consider the loss in consistency and accuracy that this more complicated action might require.

don

[don_budge]

The Role of the Legs...or...Wawrinka's inability to get a first serve in

What if one role of the legs...in addition to providing a solid, balanced platform to swing from...is to help in the propelling upwards movement (eventually through the chain of events) into the racquet head which helps to generate spin.

Wawrinka was having a monumental struggle with his first serve yesterday and if there was one thing that one could identify as the culprit of his eventual demise it was the inability to consistently get his first serve in. He has very little leg engagement from what I could tell.

He was consistently missing his first serve long which may indicate that there was not enough spin being applied to the ball. Magnus Norman...are you listening?