The Twist serve and its TWO Axis of spin. Debate Please

[Puma]

I read some posts about the third spin component of the Twist serve. I disagree with how the concept of longitudinal spin has been presented so I decided to start a new thread to discuss my theory on this.


I support the two spin idea and here is why: From a Right Hander


First lets identify the two main axis the ball spins on. One, a horizontal axis (topspin component), and Two, the angled axis (45 degree?) the sidespin component. I assume this is relatively agreed upon by us all.
With a Twist serve the ball is struck and it has two spins on it. The horizontal axis spin makes it curve downward and the sidespin makes it curve to the left like a slice spin does. What is important to realize is that the axis of spin in relation to the court change as the ball flies and curves due to its spin. What was initially a horizontal axis becomes somewhat angled and the angled side spin axis becomes near vertical or somewhat close at or near the point in which the ball hits the court. This is what gives the ball such an abrupt "kick" to the side (right).

Initially one spin (topspin) was working parallel to the side fence. The second spin (angled sidespin) was working at an angle toward the side fence. After the ball curves (later in flight) these two angles of axis change and become more angled toward the fence than before.
It is the curvature of the balls flight that changes the axis of spin on the ball that causes the sideways signature kick of the American Twist Serve.

A ball can be hit with a pure 7 to 3 angle of attack and have only 7 to 3 spin on it. This ball will curve to the left and to a degree have an enhanced downward flight. And it will bounce to the right some. But, it will not have that abrupt right hand turn that the Twist does. It is the combination of the 6 to 12 spin and 7 to 3 spin that causes the change in the flight pattern that produces that signature Twist action.

For what its worth, if you can throw very well, you can throw a tennis ball pretty easily with that 7 to 3 spin an it will follow what I described above.

Based on what I have said here, I do not believe that there is a third dimensional spin associated with the Twist. The only longitudinal turn is slight, maybe 1/8 of a turn.

Ok, hammer time. Blast my idea if you will......

 

[Kevo]

There is one spin. The axis of rotation is tilted. This makes the ball curve, but when it hits it bounces the other way. It's similar to how you can put underspin on the ball and have it bounce back toward you. I think you are over-complicating it with the multiple spins thing. I can't imagine any way to get more than one axis of rotation with a single contact. If you can, please describe it.

Do you think it takes two axes of spin to make a topspin groundstroke bounce sideways? I hit these a lot when the ball is too close to my body and I have to fight it off by pulling the racquet sideways as I make contact.

 

[AlphaCDjkr]

I read some posts about the third spin component of the Twist serve. I disagree with how the concept of longitudinal spin has been presented so I decided to start a new thread to discuss my theory on this.


I support the two spin idea and here is why: From a Right Hander


First lets identify the two main axis the ball spins on. One, a horizontal axis (topspin component), and Two, the angled axis (45 degree?) the sidespin component. I assume this is relatively agreed upon by us all.
With a Twist serve the ball is struck and it has two spins on it. The horizontal axis spin makes it curve downward and the sidespin makes it curve to the left like a slice spin does. What is important to realize is that the axis of spin in relation to the court change as the ball flies and curves due to its spin. What was initially a horizontal axis becomes somewhat angled and the angled side spin axis becomes near vertical or somewhat close at or near the point in which the ball hits the court. This is what gives the ball such an abrupt "kick" to the side (right).

Initially one spin (topspin) was working parallel to the side fence. The second spin (angled sidespin) was working at an angle toward the side fence. After the ball curves (later in flight) these two angles of axis change and become more angled toward the fence than before.
It is the curvature of the balls flight that changes the axis of spin on the ball that causes the sideways signature kick of the American Twist Serve.

A ball can be hit with a pure 7 to 3 angle of attack and have only 7 to 3 spin on it. This ball will curve to the left and to a degree have an enhanced downward flight. And it will bounce to the right some. But, it will not have that abrupt right hand turn that the Twist does. It is the combination of the 6 to 12 spin and 7 to 3 spin that causes the change in the flight pattern that produces that signature Twist action.

For what its worth, if you can throw very well, you can throw a tennis ball pretty easily with that 7 to 3 spin an it will follow what I described above.

Based on what I have said here, I do not believe that there is a third dimensional spin associated with the Twist. The only longitudinal turn is slight, maybe 1/8 of a turn.

Ok, hammer time. Blast my idea if you will......

Unfortunately, I totally disagree. I still believe that the American Twist serve maintains 3 axes of rotation. I even took the time to create a tiny project out of sheer boredom!

X axis- Topspin/Backspin
Y axis- Sidespin
Z axis- Longitudinal Sidespin.

Now look here. When you earlier stated that...

First lets identify the two main axis the ball spins on. One, a horizontal axis (topspin component), and Two, the angled axis (45 degree?) the sidespin component. I assume this is relatively agreed upon by us all.

45 degree, angled sidespin? Don't you mean a mixture between the X and Y axis, Topspin and sidespin?

After all, THIS is topspin. X-axis..

And THIS is sidespin. Y-axis..

So what is "Angled sidespin"? Is it this by any chance?

Your "Angled Sidespin" is a hybrid between the X and Y axis, Topspin-slice.

When we hit the Twist serve, we do NOT create the kind of spin shown above. No, we create something more like THIS.

As you see above, the picture displays the typical twist; Point of Contact is on the ball on the left plane, then a topspin-slice is imparted upwards and across. However, you would merely say this is Topspin-slice on a crooked axis. Now look in the background, I drew a little diagram (out of boredom! ) of a bullseye. Aye, I'm a cheap one. How can you explain the crooked spin of the twist serve using only two axes? I see three. There is topspin, right-handed sidespin, and clockwise longitudinal rotation. If the ball was really flying at the target, you'd find that the diagram is unexplainable with only two axes.
---
Another thing that assures me that there is really three axes is the twist effect. What is it that causes the sideways jolt when you use an underhanded sidespin slice? Is it topspin/backspin? Definitely not. It's not regular sidespin either, or else the ball would just curve, bounce&skid, and curve some more. It is the longitudinal spin that creates the jolt; The Z-axis spin. It is IMPOSSIBLE to create a break without Z-axis spin, and the same should apply to serves as well.

...
Hm, that's the end of my fun. This is just my view on it though, if it turns out I'm wrong I'm sorry. Haha.

 

[split-step]

I am with Kevo on this one. There is only one axis of rotation.
The way I hit it, the axis is closer to the horizontal than a 45 degree deviation. I want to have more topspin component than sidespin and 45 degree will give you equal amount of both.
Axis is also tilted into the court some. Angle of attack is from outside of ball.

I definitely don't think of all this when I hit the serve though.
A simple 7 - 1 oclock path with proper toss does the job.

 

[RestockingTues]

Unfortunately, I totally disagree. I still believe that the American Twist serve maintains 3 axes of rotation. I even took the time to create a tiny project out of sheer boredom!

X axis- Topspin/Backspin
Y axis- Sidespin
Z axis- Longitudinal Sidespin.

Now look here. When you earlier stated that...


45 degree, angled sidespin? Don't you mean a mixture between the X and Y axis, Topspin and sidespin?

After all, THIS is topspin. X-axis..

And THIS is sidespin. Y-axis..

So what is "Angled sidespin"? Is it this by any chance?

Your "Angled Sidespin" is a hybrid between the X and Y axis, Topspin-slice.

When we hit the Twist serve, we do NOT create the kind of spin shown above. No, we create something more like THIS.

As you see above, the picture displays the typical twist; Point of Contact is on the ball on the left plane, then a topspin-slice is imparted upwards and across. However, you would merely say this is Topspin-slice on a crooked axis. Now look in the background, I drew a little diagram (out of boredom! ) of a bullseye. Aye, I'm a cheap one. How can you explain the crooked spin of the twist serve using only two axes? I see three. There is topspin, right-handed sidespin, and clockwise longitudinal rotation. If the ball was really flying at the target, you'd find that the diagram is unexplainable with only two axes.
---
Another thing that assures me that there is really three axes is the twist effect. What is it that causes the sideways jolt when you use an underhanded sidespin slice? Is it topspin/backspin? Definitely not. It's not regular sidespin either, or else the ball would just curve, bounce&skid, and curve some more. It is the longitudinal spin that creates the jolt; The Z-axis spin. It is IMPOSSIBLE to create a break without Z-axis spin, and the same should apply to serves as well.

...
Hm, that's the end of my fun. This is just my view on it though, if it turns out I'm wrong I'm sorry. Haha.

Great post!

 

[split-step]

Unfortunately, I totally disagree. I still believe that the American Twist serve maintains 3 axes of rotation. I even took the time to create a tiny project out of sheer boredom!

I don't see how it is physically possible to impart more than one axis of rotation to a tennis ball with racquet. (Maybe someone can shed more light on this)

Alpha, I think you may be mixing up, axis of rotation and 'components' of the axis.


Each axis has 3 components. The x, y and z. or i, j and k.
(I really dislike physics so physics experts pls correct me if I'm wrong)

The pure slice has a purely vertical axis. Perpendicular to the baseline. It has component moving in +i and +k direction. 0 for j component. (now pure slice serve is not practical in real life and there will be some j component).

Pure topspin has horizontal axis parallel to baseline. It has components moving in +j and +k direction. 0 for i component.

topspin slice ideally has axis of 45degree to baseline. It has all 3 component, ideally with i = j component.

twist has axis of approx 10 to 30 degree to baseline. has all 3 component with j > i component.
axis is also not parallel to baseline on k component but at an angle (i.e axis intersects baseline due to angle of attack, due to toss)

 

[AlphaCDjkr]

I don't see how it is physically possible to impart more than one axis of rotation to a tennis ball with racquet. (Maybe someone can shed more light on this)

Alpha, I think you may be mixing up, axis of rotation and 'components' of the axis.


Each axis has 3 components. The x, y and z. or i, j and k.
(I really dislike physics so physics experts pls correct me if I'm wrong)

The pure slice has a purely vertical axis. Perpendicular to the baseline. It has component moving in +i and +k direction. 0 for j component. (now pure slice serve is not practical in real life and there will be some j component).

Pure topspin has horizontal axis parallel to baseline. It has components moving in +j and +k direction. 0 for i component.

topspin slice ideally has axis of 45degree to baseline. It has all 3 component, ideally with i = j component.

twist has axis of approx 10 to 30 degree to baseline. has all 3 component with j > i component.
axis is also not parallel to baseline on k component but at an angle (i.e axis intersects baseline due to angle of attack, due to toss)

Um... I, J, and K? When were the 3 dimensions of the world characterized by i,j, and k? I'm sorry, when you use these 3 letters instead of x,y, and z which I'm much more familiar with, my head starts pounding.

I'm going to get back to this post. Right now, I'm going to focus on finishing my homework and translating your post into language more suitable for my brain

 

[strahan918]

Um... I, J, and K? When were the 3 dimensions of the world characterized by i,j, and k? I'm sorry, when you use these 3 letters instead of x,y, and z which I'm much more familiar with, my head starts pounding.

I'm going to get back to this post. Right now, I'm going to focus on finishing my homework and translating your post into language more suitable for my brain

i, j, k are vector notations, if I'm right.

 

[AlphaCDjkr]

i, j, k are vector notations, if I'm right.

Vectors! Stuff that I hated in physics!

I don't see how it is physically possible to impart more than one axis of rotation to a tennis ball with racquet. (Maybe someone can shed more light on this)

Alpha, I think you may be mixing up, axis of rotation and 'components' of the axis.


Each axis has 3 components. The x, y and z. or i, j and k.
(I really dislike physics so physics experts pls correct me if I'm wrong)

The pure slice has a purely vertical axis. Perpendicular to the baseline. It has component moving in +i and +k direction. 0 for j component. (now pure slice serve is not practical in real life and there will be some j component).

Pure topspin has horizontal axis parallel to baseline. It has components moving in +j and +k direction. 0 for i component.

topspin slice ideally has axis of 45degree to baseline. It has all 3 component, ideally with i = j component.

twist has axis of approx 10 to 30 degree to baseline. has all 3 component with j > i component.
axis is also not parallel to baseline on k component but at an angle (i.e axis intersects baseline due to angle of attack, due to toss)

Ok, after figuring out the little i/j/k mystery thanks to strahan, I must say... Um.. I still don't have much to say. I just took a quick shower and I feel much better now, and I thought about it a little more.

So, I created a little.. comparison!

We all understand that we live in a 3D world. 3 Dimensions, 3 Axes, X/Y/Z (or i,j, and k but I wont refer to them as these, ok?).
When we walk along the X-Axis, we move left and right. How about, this is Width.
When we walk along the Y-Axis, we move up and down (jumping, changing ALTITUDE.) How about this be, Length?
When we walk along the Z-Axis, we move forward and backward. Lets call this Depth.

Length, Width, Depth? We practically live in a BOX. The earth, our WORLD is a BOX (Fine, sphere to represent the round earth.) We are the little people traversing the insides of this BOX in the 3 dimensions. Now, we are all standing on the ground, aren't we? When we take a step forward, we traverse the Z axis. When we sidestep to our left, we go on the X axis. When we jump, we go up on the Y axis. So what happens when we jump diagonally left/forward? We did NOT travel on one axis right now. There is no axis in our definitive world that goes at a diagonal tangent upwards and to our left.

No, we just traveled some distance on all three axes at the same time. The same should apply to the tennis ball. The ball can utilize a X axis spin, topspin, a Y axis spin, sidespin, and and Z axis spin, Longitudinal sidespin. So logically, when we hit a ball with a mix of Topspin and Sidespin, we did NOT hit the ball along an axis between the X and Y axes; There IS NO SUCH AXIS squished between the X and Y axes! We can only say we imparted both X and Y spin at the same time; 2 axes. After all, last time I did a reality check, moving diagonally was not an axis on its own. Do we live in a 4D world, where 2 of the dimensions coexist to form a kind of 3rd hybrid axis? Naw.

Also Split-step, I'm sorry but I'm really not in the mood to discuss the majority of your post... with the i j k stuff. My AP Physics class never discussed vectors too much; maybe its a collegiate subject? I'm really not too happy to have to think about the angles and letters you've given me. So, I'm not going to reply to that part; sorry.

Ok then, sleep time.

 

[Raidenx]

A freely rotating ball cannot be rotating about more than one axis at an instant no matter which way it spins. It's all about how the axis of rotation is oriented.

You can however say a ball has two or three components of spin of varying magnitude based on some reference coordinate system, which you could superposition together to obtain the resultant spin direction.

 

[SystemicAnomaly]

I'm turning myself in -- I am one of the heretics or blasphemers, if you will, that have been promoting the notion of 3rd axis spin on twist serves and some underarm groundstrokes in the past couple of months. This 3rd axis spin is akin to Roll in flight dynamics. For tennis & other sports we can refer to it as longitudinal spin, spiral spin, or spiralspin. The notion of spiral spin in tennis occurred to me several years ago but I never really gave it a great deal of thought until this past Spring when I read Kathy Krajco's articles on spin doctoring and on the twist serve on her Operation Doubles web site.

.....

..... Axes of Rotations
Topspin-Slice ..... Twist Serve

AlphaCDjkr has done a fine job of describing and illustrating what is happening with spiral spin and other spins. As others have pointed out, a ball in flight really only has one spin or one axis of rotation. However, that spin can consist of anywhere from 1 to 3 components. For the sake of simplicity, I prefer to stick with the referring to the reference axes as X, Y, & Z rather than using the vector notation. (We could also call those horizontal, vertical, and longitudinal axes, respectively).

Note that the sidespin component is about the vertical axis, not an angled axis as Puma indicates. The angled axis that Puma refers to would have both a horizontal (topspin) and a vertical (sidespin) component as seen in a topspin-slice serve. For conventional spins in tennis the axis of rotation is always perpendicular to the flight of the ball. This is true for topspin, backspin, and sidespin (both left & right flavors) and combinations of sidespin and either of the horizontal spins.

For spiralspin, the axis of rotation is parallel to the flight of the ball. According to Operation Doubles and other sources, the axis of rotation for a Twist serve is an offset or twisted axis -- it is not perpendicular to the flight of the ball as with other types of serves. This can only happen if there is some 3rd axis spin present in the Twist serve. If no spiralspin is present, then would would have a regular topspin-slice serve. It is the longitudinal component, the spiralspin, that accounts for the difference between these 2 serves.

Note that horizontal spins (topspin & backspin) and vertical spins (left & right sidespins) cause the ball to travel in a curved path due to the Magnus effect. Spiral spin, on the other hand, does not cause the ball to curve since it does not produce the Magnus effect.

However, the court interactions are different for these various types of spins. While a sidespin will cause a deviation when the ball travels thru the air, it does not result in a kick or left/right deflection when the ball bounces. On the other hand, a spiral spin causes no deviation thru the air but does cause the ball to kick to the right or the left when it interacts with the ground.

After doing some additional research on the subject today, I have discovered that physicist Rod Cross (PhD) and Crawford Lindsay have talked about spiralspin in length in their 2005 book, Technical Tennis. They discuss this 3rd axis spin in Chapter 4, Spin & Trajectory -- on pages 124 thru 128.
.

 

[SystemicAnomaly]

^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^

<--- This is the book referenced in my post above.

To see the effects of various spins, try some simple throws with a tennis ball. Throw the ball forward with a vertical axis spin (left sidespin or right sidespin). Note that the ball does not kick to the right or the left when it bounces. Now try a forward throw imparting a spiral spin -- if you give it a clockwise spiral spin, the ball kicks off to the right; if your impart a counter-clockwise spiral spin, it bounces off to the left.

Also, try the various spins without a forward throw. Drop the ball with a forward (top) spin. The ball bounces away from you, while the reverse spin will bounce toward you. A clockwise spiral spin will bounce bounce to the right while a CCW spiral spin will bounce left. Now drop the ball with spin about the vertical axis -- the ball should bounce straight up (as long as the drop really is vertical and no X or Z spin is present). This should happen whether you impart a left sidespin or a right sidespin.
.

 

[SystemicAnomaly]

This image is Figure 4.4 on page 127 of Technical Tennis. It illustrates 6 of the 18 different spin combinations possible in tennis. (The other 12 combinations would include Backspin as well as counter-clockwise Spiralspin and the other flavor of Sidespin).

 

[jrod]

An Idea for Hawkeye?

I like SA's description here....the one thing that isn't discussed clearly is the ball's relationship to the reference axes as it moves through space. The axes are static and do not change but the ball's rotation about these axes is not static and can change as a result of the force and spin imparted on the ball and pressure differentials created by these forces.

Occassionally on Tennis Channel they use a graphic comparing the flight path of the serve of the two players. What they do not show is the spin component of the ball as it traverses the flight path. If you can imagine breaking down the flight path into a sequence of snapshots (e.g. the frame rate of Hawkeye system that is used to record flight path) and superimposing a vector (e.g. a pencil) showing the axis of rotation at each point along the flight path, then you would see the dynamic I am refering to.

I vote for SA to call the inventors of Hawkeye and suggest they add this information to their graphical toolchest.

 

[Puma]

Alpha & Systemic,

Thanks for your reply.

Alpha, great pictures by the way.

I am speaking of the angled sidespin indicated in the last picture you present. The results of striking the ball with an upward blow moving from say 8 to 2 on the ball.

I do not buy the idea that there is a longitudenal spin on the ball. There is nothing in the process of striking the ball that would impart such a spin.

It is the fact that the ball bends or curves in flight, which means that what once was pure topspin x axis ends up not being in same orientation when the ball hits the court as it was when it left the racket. The sam is true for the z axis. As the ball curves or bends in flight the z axis is more oriented to the side fence, hence the abrupt right hand kick of the serve.


I do understand and realize what you are saying about a 3d world and such. However, I don't believe there is a third spin associated with this serve.

Nice debate huh???

 

[jrod]

.....It is the fact that the ball bends or curves in flight, which means that what once was pure topspin x axis ends up not being in same orientation when the ball hits the court as it was when it left the racket. The sam is true for the z axis. As the ball curves or bends in flight the z axis is more oriented to the side fence, hence the abrupt right hand kick of the serve....

Like I stated above, the ball's orientation in flight including it's axis of rotation are dynamic, not static. I still think this can be illustrated best using Hawkeye projections of the ball's flight. If you can visualize just after the point of impact with say a pencil penetrating the ball on it's axis of rotation. The pencil's orientation will change as the ball moves through space. A movie of how the pencil rotates through the flight path would illustrate this quite well.

 

[Puma]

Note that the sidespin component is about the vertical axis, not an angled axis as Puma indicates. The angled axis that Puma refers to would have both a horizontal (topspin) and a vertical (sidespin) component as seen in a topspin-slice serve. For conventional spins in tennis the axis of rotation is always perpendicular to the flight of the ball. This is true for topspin, backspin, and sidespin (both left & right flavors) and combinations of sidespin and either of the horizontal spins. This may be true for the purposes of explanation. However, in reality this is not true at all. I referred to the angled side spin as angled side spin because that is what it is. A ball can rotate on whatever axis that it was put into motion under. Whether it is perpendicular to its flight path has no bearing. The paragraph below indicates this. You can define things any which way. I defined things in my own way. Sorry for that. But i did not refer to side spin as being on a verticle axis.

For spiralspin, the axis of rotation is parallel to the flight of the ball. According to Operation Doubles and other sources, the axis of rotation for a Twist serve is an offset or twisted axis -- it is not perpendicular to the flight of the ball as with other types of serves. This can only happen if there is some 3rd axis spin present in the Twist serve. If no spiralspin is present, then would would have a regular topspin-slice serve. It is the longitudinal component, the spiralspin, that accounts for the difference between these 2 serves This is false. It is the combination of the two spins that creates its flight pattern that results in the balls orientation to the court once it hits the court that causes the twist kick action.

What I tried to define is exactly what was in the last picture. These axises of rotation create a flight pattern in which the ball curves and dips at the same time. Thus the axis of spin "Change" in relation to what we realize is up/down/sideways to the court surface.

 

[Puma]

Sorry for so many posts. I get timed out if I wait too long.


I would go on to say that it is specifically the "speed" of rotation of these two spins that makes the flight characteristics thus the bounce characteristics unique.

Also, I am not speaking of a topspin slice serve. I know what that is. A person can throw a tennis ball that mimics this flight pattern. A ball only needs on axis of rotation in order to fly and bounce as it does.

 

[SystemicAnomaly]

Puma, I don't quite get why you don't accept the notion of spiralspin (z axis rotation). I'll have to review your posts again later to see if I can figure out where the problem lies.

If you have actually tried the throwing & dropping exercises that I proposed on post #12, it should become readily apparent that longitudinal (spiral) spin must account for the side bounce seen with a twist serve since the topspin and sidespin components do not result in this type of ground interaction effect.

Topspin & sidespin does account for the curved trajectory effect, while z-axis spin does not. However, these Magnus effect manifestations (due to air pressure differentials) should not be confused with the different ground interactions seen with different spins.

Look again at pictures that illustrate the differences between topspin-slice and twist serves -- pictures 3 & 4 in post #3 and the bottom 2 graphics in post #13. Comparisons of these pictures should show how spiralspin is imparted in one case and not the other. For the topspin-slice serve, the brushing action is perpendicular to the radius of the ball (at a "point" on the back of the ball). For the twist serve, the brushing is off-center -- not perpendicular to the radius of the ball. This carving of the ball is what produces the spiralspinon the twist serve.

Will look at your last 3 posts later & try to glean what you are trying to say.
.

 

[Puma]

Puma, I don't quite get why you don't accept the notion of spiralspin (z axis rotation). I'll have to review your posts again later to see if I can figure out where the problem lies.

If you have actually tried the throwing & dropping exercises that I proposed on post #12, it should become readily apparent that longitudinal (spiral) spin must account for the side bounce seen with a twist serve since the topspin and sidespin components do not result in this type of ground interaction effect.

Agreed. However, you are missing the point I am raising. The topspin and side spin mentioned are referred respectively topspin and side spin this because of its orientation upon leaving the racket. Once the ball is in flight it curves as we know. The effects of the curvature in flight causes the orientation of the ball to change. That is, what was once the top of the ball (12:00) is more toward say 1:00 or more because of the curve of the flight. Thus the side spin has changed was well. Once the ball hits th court the side spin component is spinning the same direction as your spiral spin. Thus it is the topspin and sidespin turned to the side somewhat that causes the sideways kick of the ball.

Topspin & sidespin does account for the curved trajectory effect, while z-axis spin does not. However, these Magnus effect manifestations (due to air pressure differentials) should not be confused with the different ground interactions seen with different spins.

Look again at pictures that illustrate the differences between topspin-slice and twist serves -- pictures 3 & 4 in post #3 and the bottom 2 graphics in post #13. Comparisons of these pictures should show how spiralspin is imparted in one case and not the other. For the topspin-slice serve, the brushing action is perpendicular to the radius of the ball (at a "point" on the back of the ball). For the twist serve, the brushing is off-center -- not perpendicular to the radius of the ball. This carving of the ball is what produces the spiralspinon the twist serve.

I understand where the ball is hit. However, I respectively disagree that sprial spin is imparted.


Will look at your last 3 posts later & try to glean what you are trying to say.
.

..........

 

[sureshs]

A rigid body (and a tennis ball is not one but for these purposes will do) can have only one axis of spin. It is a different matter that the resultant motion can be mathematically described as the combined effect of separate spins around 3 perpendicular axes.

 

[CoachingMastery]

A ball in flight can't have more than one axis of spin once it initiates its movement. Unless an outside force acts on the ball, it will continue to rotate on whatever axis it starts on.

An airplane can have several axis of movements while in flight because of the simple aspect of flaps and rudder which can create a change in a planes movement.

A player can manipulate the axis relative to the flight of the ball with a specific swing path or contact point or both.

Remember too, that the axis can be vertical, horizontal or degrees of both; in addition, the horizontal axis can be parallel to the baseline or pointed at an angle (ie: towards a net post). These combinations of axis is what determines the type of spin: slice (vertical axis for a true slice); hybrid (about 45 degree tilt to the left for a right handed player...still with the axis parallel with the baseline); Topspin (horizontal axis parallel with the baseline for a true topspin serve); Kick (think reverse hybrid: axis that is somewhat at a 45 degree angle, but the top point of the axis is pointed up to the right for a right handed player...also the axis is parallel to the baseline within this tilt); twist (same as a kick except except instead of the axis lying parallel to the baseline, the bottom point of the axis is pointed more towards the left net post, with accent on contact being more up the inside portion of the ball...ie; hitting up the side of the ball closest to the server.)

These are the generally accepted serves that top players use. Obviously, we can create a complete "back spin" serve, by hitting completely under the ball or have the racquet sliding down at contac, (by pulling the hitting elbow down)...a serve that many beginner players use, as well as those who did not learn to serve correctly! Obviously, this serve has severe limitations in terms of spin, due to the underspin component.

Good diagrams by those who included them. They will help others see these spins.

 

[AlphaCDjkr]

Just got back from school, to realize...

That I see I've confused myself thoroughly while I tried to show my side of the argument.

Yes, I think I need to make it clear to myself and others that I understand that a ball spins on only one axis. However, my argument remains the same, and now I will kind of fix it to state what I originally wanted to say.

I meant, that though a ball does spin on one axis only, my view is that that axis in relation to the trajectory/angle to the court and flight path, is composed of all three, X/Y/Z, components of rotation, not just topspin and sidespin.

My apologies to Kevo, split-step, Raidenx, sureshs, and CoachingMastery for my own confusion on the topic.

 

[AlphaCDjkr]

Ugg.. Puma, why did you quote Systemic by typing in his quote, along with making the text red? Made it blinding to quote and read D:

Here's my view on things. You don't believe that you can exhibit Z-axis spin on the ball. That's fine with me, I'll attempt to show why I believe that you can.

Can we all agree that a twist serve is essentially, a Topspin-slice serve hit on the left face of the ball with a crooked axis of rotation? After all, if it was really just a topspin-slice serve hit on the left side of the ball, the ball would behave like a regular, non-twisty, topspin-slice serve just aimed further right.

No however, the twist serve is NOT simply a topspin-slice hit on the left face of the ball. When we first contact the ball, we proceed to pronate up and around the top/front of the ball- The SPIN is therefore a traditional topspin-slice; however, the swingpath/pronation/followthrough is what creates the crooked axis of rotation, along with the redirection of the serve to go INTO the court, rather than shooting off to our right.

Logically, with this information, the ball WOULD only impart X and Y spin; Topspin and sidespin. However, that is assuming we look straight through the point of contact of the ball towards the target. For a twist serve, the Point of Contact of the ball does NOT equal the direction of the ball traveled.

You can see what I mean here...

Ok, there's our twist serve. Our point of contact is always at the left of the ball; judging by that, wouldn't you agree that the flight path of the ball would NOT be the target I drew there, but would maybe be off to the right? However, our twist serve DOES go in the direction of the target. Now, how could that be?

It's the twisted axis here. When we realign our perspective from the ball -> target, that is an unidentifiable spin. It's not traditional, and we rarely see it. What kind of ball spins in a way that doesn't match it's flightpath? No ball does, with the exception of twist serves, underhanded spiral-sidespin, and various other trickshots. By realigning our perspective, we can see that the ball indeed contains all 3 components of spin, resulting in the crooked axis.

...
Great debate yeah.

 

[Solat]

...
Great debate yeah.

would you call it a mass debate? :mrgreen:

 

[AlphaCDjkr]

would you call it a mass debate? :mrgreen:

Is it? I don't think so

I think... it's more like a heated argument between conflicting views for the good of common sense and knowledge for the universe. The outcome will decide the fate of all tennis balls in the world. Spin or no spin? Twist or no twist?

One ball to rule them all, one thread to find them.
One post to bring them all and...
...
...
...
wat?

 

[SystemicAnomaly]

Note that the sidespin component is about the vertical axis, not an angled axis as Puma indicates. The angled axis that Puma refers to would have both a horizontal (topspin) and a vertical (sidespin) component as seen in a topspin-slice serve. For conventional spins in tennis the axis of rotation is always perpendicular to the flight of the ball. This is true for topspin, backspin, and sidespin (both left & right flavors) and combinations of sidespin and either of the horizontal spins...

This may be true for the purposes of explanation. However, in reality this is not true at all. I referred to the angled side spin as angled side spin because that is what it is. A ball can rotate on whatever axis that it was put into motion under. Whether it is perpendicular to its flight path has no bearing. The paragraph below indicates this. You can define things any which way. I defined things in my own way. Sorry for that. But i did not refer to side spin as being on a verticle axis...

Let's stick with the conventions. If we all start making up our own definitions it creates confusion and makes it nearly impossible to carry on a discussion. By definition, sidespin is spin about a vertical axis. Topspin & Backspin are spins about a horizontal axis that is perpendicular to the direction of travel. Take a look at the Op Doubles link I provided (in post #11) for these definitions of the Axes of Rotation. The perpendicular aspect cannot be ignored -- it is germane to the definition. This right angle relationship is also important because it tells us if the Magnus Effect is present for a given type of spin.

Note that spiralspin, by definition, is also on a horizontal axis. However, it is parallel to the direction of travel & therefore does not result in the Magnus Effect. (Take a look at Technical Tennis by Cross & Lindsey for these facts).

For spiralspin, the axis of rotation is parallel to the flight of the ball. According to Operation Doubles and other sources, the axis of rotation for a Twist serve is an offset or twisted axis -- it is not perpendicular to the flight of the ball as with other types of serves. This can only happen if there is some 3rd axis spin present in the Twist serve. If no spiralspin is present, then would would have a regular topspin-slice serve. It is the longitudinal component, the spiralspin, that accounts for the difference between these 2 serves.

This is false. It is the combination of the two spins that creates its flight pattern that results in the balls orientation to the court once it hits the court that causes the twist kick action...

No, this is not false. Take a look at both Operation Doubles and Technical Tennis if you doubt what I am stating here.

What I tried to define is exactly what was in the last picture. These axises of rotation create a flight pattern in which the ball curves and dips at the same time. Thus the axis of spin "Change" in relation to what we realize is up/down/sideways to the court surface.

Yes, the ball can curve (left, right, up, down) during its flight, but the axis of spin maintains the same relationship to the direction of flight (prior to the bounce) at any point in its trajectory. I believe that it probably also maintains the same relationship to the court surface. It is gravity and the Magnus Effect due to the presence of topspin/underspin and sidespin that is responsible for the flight pattern (trajectory).

.

 

[SystemicAnomaly]

...

I do not buy the idea that there is a longitudenal spin on the ball. There is nothing in the process of striking the ball that would impart such a spin...

The manner in which the ball is stuck has everything to do with whether or not z-axis spin is present. If we brush directly across the back of the ball at some angle we will get some combination of topspin/backspin and sidespin. However, it we carve around the ball so that we contact it off-center, we can also impart spiralspin.

To clearly see spiral spin in action on a tennis ball, try the following. Toss a ball up in front of you. As the ball descends, brush across the bottom of the ball from left to right (or right to left). The spin imparted is spiral spin. When the ball bounces it will kick off to the side. Now do the same thing, but brush the ball a somewhat above the very bottom (but not directly on the back of the ball). Stroke the ball as if you were drawing a smile on the lower part of the ball. by doing so, you will still impart spiral spin but will also probably impart some sidespin &/or underspin. If sidespin is present, the ball will curve in flight. But it will still be the spiral spin that is responsible for the kick off to the side when the ball bounces.

The spiral spin is imparted because we did not brush directly on the back of the ball. The same thing happens on a twist serve because we are not brushing directly on the back of the ball. However, in this case we have imparted topspin rather than underspin. But the principle is the same.

Note that Technical Tennis (by Cross & Lindsey) appears to back up everything that I've observed about spiralspin and other types of spins. I strongly urge you to get your hand on it and give Chapter 4 a thorough reading to see what these tennis researcher/scientists have to say about spiralspin & other types of spins.

I just read an interesting detail in Technical Tennis about the relationship between sidespin and spiralspin on serves. The following quote is from page 125 in the section on Spiralspin:

"... the usual situation is when a player serves a ball so that it curves, say, to the left through the air, then it will straighten up when it bounces by kicking to the right. That's because the action of slicing the ball so it spins about a vertical axis also causes the ball to spin about a horizontal axis like a football. You can observe the action best if you stand behind a player when he or she is serving... "

This excerpt indicates that there will be some component of spiral spin on slice serves that are not necessarily twist serves. It has to do with the manner in which slice is imparted to the ball on a serve. This spiral component is mild compared to that of a twist serve. As a result, the ball direction straightens when it bounces rather than having a pronounced kick to the side (as seen with the twist serve).
.

 

[jrod]

....Yes, the ball can curve (left, right, up, down) during its flight, but the axis of spin maintains the same relationship to the direction of flight (prior to the bounce) at any point in its trajectory. I believe that it probably also maintains the same relationship to the court surface. It is gravity and the Magnus Effect due to the presence of topspin/underspin and sidespin that is responsible for the flight pattern (trajectory).

SA- I'm not so sure about this statement. It seems to me that the axis can move as well during flight. It might not move much, but I'd bet it changes orientation in some cases. Certainly after contact the ball's axis of rotation can change.

 

[BrianGordon]

I don't think this is complicated enough - let's add a fourth dimension and consider the quaternion representation of the rotation.

 

[jrod]

I don't think this is complicated enough - let's add a fourth dimension and consider the quaternion representation of the rotation.

Agree...by the way, I checked out your site. That's some pretty fancy gear you are using to study players. Don't you have an issue with the Heisenberg uncertainty principle when you use all that gear? Is wireless sensor technology not mature enough to use for this purpose?

 

[BrianGordon]

Agree...by the way, I checked out your site. That's some pretty fancy gear you are using to study players. Don't you have an issue with the Heisenberg uncertainty principle when you use all that gear? Is wireless sensor technology not mature enough to use for this purpose?

Have not heard that one in a while - since the effect is really only measurable on a subatomic scale, we don't worry about it too much - there are bigger problems to deal with. The transmitter pack for the wireless system is actually more cumbersome than the cables so I felt wireless did not justify the nearly doubling of the cost - anyway, I'm moving to a real time optically based LED recognition system with 480 hz speed, 1/10 th mm accuracy, and minimal stroke interference this Spring.

 

[sureshs]

Don't confuse rotation about an axis and rotation of the axis. For example, airplanes cannot turn like cars in flight. When they turn right or left, they also tilt downwards on the side in the direction of the turn. It can be looked at as rotation about one axis (turning) while the axis itself it rotating about another axis perpendicular to it (tilting).

 

[jrod]

Have not heard that one in a while - since the effect is really only measurable on a subatomic scale, we don't worry about it too much - there are bigger problems to deal with.

Wow, tell me you weren't serious here? All I was trying to say was that it seemed to me the weight and cabling may actually interfere or distort the data you are trying to collect in assessing a players stroke mechanics. It certainly would bother me.

The transmitter pack for the wireless system is actually more cumbersome than the cables so I felt wireless did not justify the nearly doubling of the cost - anyway, I'm moving to a real time optically based LED recognition system with 480 hz speed, 1/10 th mm accuracy, and minimal stroke interference this Spring.

Interesting feedback on wireless systems. I'm a little surprised to hear you say the Tx pack is cumbersome. I have a continuous glucose monitoring system that is wireless and the Tx is not at all cumbersome. What is the data rate you need to transmit? The range is certainly less than 10 feet.

Also, is your system collecting any data that cannot be gleaned from high speed video? Have you looked into what data could be collected using Wii technology? Just curious...

 

[Steady Eddy]

For the topspin-slice serve, the brushing action is perpendicular to the radius of the ball (at a "point" on the back of the ball). For the twist serve, the brushing is off-center -- not perpendicular to the radius of the ball..

What do you mean by "the radius of the ball"? In a sphere, or ball, any line from the center to the surface is the radius. Using the word "the" implies that there is only one. Furthermore, it is impossible not to hit the ball at a point perpendicular to the radius. I've taught Geometry, so I understand the terms. There's only one spin on the ball as it travels over the court. Even if it was spun at a 45 degree angle, so that it has both topspin and sidespin the ball is only spinning in one way.

 

[jrod]

What do you mean by "the radius of the ball"? In a sphere, or ball, any line from the center to the surface is the radius. Using the word "the" implies that there is only one. Furthermore, it is impossible not to hit the ball at a point perpendicular to the radius. I've taught Geometry, so I understand the terms. There's only one spin on the ball as it travels over the court. Even if it was spun at a 45 degree angle, so that it has both topspin and sidespin the ball is only spinning in one way.

Right....now, freeze the axis of rotation at various points in the balls flight and examine the behaviour of the axis of rotation. I contend that relative to the reference coordinate system, the axis of rotation probably does not remain fixed throughout the flight of the ball. Any presssure differentials created by a spinning ball will cause the ball to curve and possibly change the orientation of the axis about which the ball is spinning.

That said, I am wondering what the hell difference this is going to make to anyone trying to impart spin on a tennis ball...

 

[Zachol82]

It is the combination of the 6 to 12 spin and 7 to 3 spin that causes the change in the flight pattern that produces that signature Twist action.

I agree with the 2-axis of spin idea. By your statement quoted above, you mean hit the ball as if it is 6:15 or maybe 7:15? Following the short hand from 6 or 7 up to the middle, then the long hand from the middle to 3 (in this case 15 minutes).

 

[BrianGordon]

Wow, tell me you weren't serious here? All I was trying to say was that it seemed to me the weight and cabling may actually interfere or distort the data you are trying to collect in assessing a players stroke mechanics. It certainly would bother me.

Interesting feedback on wireless systems. I'm a little surprised to hear you say the Tx pack is cumbersome. I have a continuous glucose monitoring system that is wireless and the Tx is not at all cumbersome. What is the data rate you need to transmit? The range is certainly less than 10 feet.

Also, is your system collecting any data that cannot be gleaned from high speed video? Have you looked into what data could be collected using Wii technology? Just curious...

Sorry- Heisenberg is a concept very specific to quantum physics - clearly did not understand your meaning. Yes the cables interfere some with the strokes - but actually very little - kids as young as 7 have no problem reproducing their normal stroke mechanics.

The wireless transmitter pack was about 5 lbs but individual sensors still had to be wired so the difference was minimal - the sensors (10) transmit translation and rotation (4x4 matrix) at 144 hz each.

We use high speed video all the time in my academic lab - the extensive post-processing (manual frame by frame digitizing of 35 body points) of at least 2 cameras to reconstruct 3D space is too time consuming - 2D is useless in my application. For player development I need real time 3D.

Not familiar with the Wii technology but in seeing it used assume it must use accelerometers - not on par with the more advanced technologies to measure full body 3D motion at high rates.

 

[SystemicAnomaly]

SA- I'm not so sure about this statement. It seems to me that the axis can move as well during flight. It might not move much, but I'd bet it changes orientation in some cases. Certainly after contact the ball's axis of rotation can change.

I'll agree with this. I don't really know if the axis of rotation wanders or shifts (like the Earth's axis does). If it does, I would suspect that it does not move all that much with respect to the trajectory of the ball prior to the bounce.

I also agree with an early statement that you made. The spin rate can change during flight, not just at the bounce. I have observed balls increasing in spin rate during flight many times. It amazes me when I actually see this happen.

 

[Puma]

Alpha,

I quoted SA within his quote by mistake. Sorry for that. I respect both you and SA for your tennis advice and comments. I am sincerely interested in this spin discussion and have no intention of this discussion becoming heated or petty. I typed in red to distinguish what I was saying in reference to what he was saying. No harm meant. I am now writing this in a word processor so I can better communicate my ideas.


SA,

Thank you for your well thought out and well researched posts. I appreciate your knowledge. In reference to your post #27 I agree with almost everything you say here except for the part where you say the spiral spin on a twist is horizontal. I can visualize and understand what you say. Please let me try to explain my claim better so you can visualize what I mean. I realize we may not agree, and that’s ok. I just want for you to be able to understand what my claim is. I am doing my very best to be clear. I do not have the books you refer to. However, I did go to the link you provided about the axis of spins on a tennis ball.
On that web page it clearly indicates the different spin for a slice serve and a twist serve. The spin on the twist serve is moving from say 7 to 1. This had no horizontal or vertical axis shown in his example. It is tilted or angled as is the topspin slice which appears to be more like from 8 to 2. Of course he references the topspin axis as horizontal and the slice spin as vertical. And I agree.

It is my claim that with a twist serve the ball spins from approx 6 to 12 & say 7 to 1 or 8 to 2 simultaneously. Two spins at once. One being topspin and the other being angled. A combination of what is shown on the web site. The result is a ball that curves both down and to the left during flight. It is due to the curvature of the balls flight that the orientation of the two spins (axis) change in reference to the court. What was once 6 to 12 becomes 7 to 1, and, what was once 7 to 1 or 8 to 2 moves over a notch to 8 to 2 or 9 to 3. Remember 8 to 2 or 9 to 3 would be really close the spiral spin you speak about. However, it is my claim that there really is no spiral spin. The action or abrupt bite that ball takes on the court is from the "skew" or the ball tilting slightly so that the two spins or pointing toward the right at impact with the court.

A topspin slice only needs the one angled spin to act the way it does. Spin from 7 to 1 or 8 to2 will do for this serve. I agree with your statement about the Magnus effect and gravity’s effect on the balls flight. I do not agree that the axis of rotation always remains the same.
SA, I appreciate your efforts in this thread. I am not trying to pick a fight, I am trying to put forth my claim as to why I think the twist acts as it does. I have no idea how I know this, aside from hitting a million golf balls and watching them hook and slice etc. This is where I learned a ball can spin two different directions at the same time.

Ok, I’m done. That’s about the best I can explain myself.

 

[SystemicAnomaly]

What do you mean by "the radius of the ball"? In a sphere, or ball, any line from the center to the surface is the radius. Using the word "the" implies that there is only one. Furthermore, it is impossible not to hit the ball at a point perpendicular to the radius. I've taught Geometry, so I understand the terms. There's only one spin on the ball as it travels over the court. Even if it was spun at a 45 degree angle, so that it has both topspin and sidespin the ball is only spinning in one way.

Thanks for pointing this out. After I wrote that, I realized that I had not explained it clearly enough. The radius (or diameter) that I was referring to is the one the that goes thru the back of the ball. Let me quote Technical Tennis (Cross & Lindsey) so that the meaning is clear:

"Spiralspin is generated when the racquet is moving across the face of the ball (upward, downward, or sideways) in such a way that the strings make contact with the ball at any point that is not exactly at the back of the ball--that is, not coincident with the spin axis."

.

 

[mucat]

You guys still debating this? This is what I draw last time and about 99.9% of the TW members disagree with me, haha. At least most of us here are not saying twist spin is just topspin + lefty sidespin, so we are making progress here.

I remember someone mentioned twist spin is a combination of x-axis spin (toppspin), y-axis spin (sidespin) and z-axis spin (clockwise spin?). Yes it is only one axis of rotation, but it is a combination of all this. Personally, I think a pure twist serve will have only topspin + clockwise spin, but I am not betting on it though, because the most important component of the twist serve is the clockwise spin.

 

[AlphaCDjkr]

You guys still debating this? This is what I draw last time and about 99.9% of the TW members disagree with me, haha. At least most of us here are not saying twist spin is just topspin + lefty sidespin, so we are making progress here.

Topspin and lefty-sidespin? Who said that? :shock:

I remember someone mentioned twist spin is a combination of x-axis spin (toppspin), y-axis spin (sidespin) and z-axis spin (clockwise spin?). Yes it is only one axis of rotation, but it is a combination of all this. Personally, I think a pure twist serve will have only topspin + clockwise spin, but I am not betting on it though, because the most important component of the twist serve is the clockwise spin.

That's what I thought. I felt that in a twist serve, the -
X-axis (Topspin) provided the consistency/kick that shows why this is a variant of the kick serve.
Y-axis (Sidespin) was responsible for any curvature in the air.
Z-axis (Logitudinal spin) was responsible for any twist effect.

Individually, each one of these axes would behave as described above- I don't feel that combining the spins would disrupt the behavior of the ball as a whole.

Well then, I could persuade myself to believe that a Perfect Twist serve would actually be 100% Z-axis spin. However, this would be impossible as there is no way to impose 100% logitudinal spin while directing the serve FORWARD; The contact point for a longitudinal spin is always at the sides/top/bottom of the ball - the places where it is impossible to impose any forward momentum.

Hm, I need to go somewhere tonight and I probably won't be back on. This is a pretty constructive argument/debate, I'd like to see where this ends up.

 

[SystemicAnomaly]

The Twist serve exhibits a curved trajectory that indicates that it must have a sidespin component. In fact, for a serve, it would be probably be impossible to generate spiralspin w/o also imparting sidespin. According to Cross & Lindsey, the opposite also appears to be true -- any slice serve will have a spiral component. However for non-Twist serves, this spiral component will be mild.

 

[Puma]

The abilitly to realize the adjustment in axis orientation completely satisfies the idea or concept of spiral (counterclockwise) spin.

The spin of the 7 to 1 or 8 to 2 brush of the ball "becomes" more spiral at impact (court) versus its orientation from impact by the racket.

 

[Bungalo Bill]

The abilitly to realize the adjustment in axis orientation completely satisfies the idea or concept of spiral (counterclockwise) spin.

The spin of the 7 to 1 or 8 to 2 brush of the ball "becomes" more spiral at impact (court) versus its orientation from impact by the racket.

:shock::shock::shock::shock:

 

[mucat]

I think we should talking about the spin in a technical and purest form. I mean, topspin groundstrokes will always have sidespin component, and the sidespin will be mild and unintentional (unless you are one of the Pros), but this is just because no one can hit a pure topspin groundstrokes in practice.

So, I don't think slice serve has the spiral spin (z-axis) component. Because spiral spin is what generated the twist. Also, I think someone already mentioned the aixs of rotation is alyways related to the direction of travel? I don't know if it is true or not. But if it is, this probably means the axis of rotation should not change...at least I can visualize the axis of rotation should not change for a slice serve.

Originally, I thought the twist serve only has topspin+spiralspin, but you guys already suggest it is impossible physically to generate topspin+spiralspin without the sidespin component, and I tend to agree with this now. However, I don't know how much is the sidespin is contributing to the action of the ball. If you look at a twist serve and a topspin-slice before the bounce, the action in the air is a bit different.

I think the physics major here should come out and give us some pointers...

 

[Bungalo Bill]

I think we should talking about the spin in a technical and purest form...So, I don't think slice serve has the spiral spin (z-axis) component. Because spiral spin is what generated the twist...I think the physics major here should come out and give us some pointers...

:shock::shock::shock::shock:, okay.

 

[mucat]

What I am trying to say is, when we are talking about topspin serve, sometimes, someone will mention there should be some sidespin in it because it is humanily impossible to produce spin with perfect x-axis rotation. However, this doesn't mean a topspin serve will have sidespin in it technically.

SystemicAnomaly mentioned there might be spiral spin component in slice serve (which I disagree). I just want to know if he means technincally or practically.

 

[SystemicAnomaly]

...

So, I don't think slice serve has the spiral spin (z-axis) component. Because spiral spin is what generated the twist. Also, I think someone already mentioned the aixs of rotation is alyways related to the direction of travel? I don't know if it is true or not. But if it is, this probably means the axis of rotation should not change...at least I can visualize the axis of rotation should not change for a slice serve.

Originally, I thought the twist serve only has topspin+spiralspin, but you guys already suggest it is impossible physically to generate topspin+spiralspin without the sidespin component, and I tend to agree with this now. However, I don't know how much is the sidespin is contributing to the action of the ball. If you look at a twist serve and a topspin-slice before the bounce, the action in the air is a bit different.

I think the physics major here should come out and give us some pointers...

If it's a physics expert that you want then pay heed to the words of Rod Cross . He and Crawford Lindsey addressed all of this stuff and a lot more in their 2005 book, Technical Tennis (take a look at page 1 of this thread for a link & several references to this book).

Also, look closely at the quote from these guys that I included in post #28 of this thread. It states, "...the action of slicing the ball so it spins about a vertical axis also causes the ball to spin about a horizontal axis like a football...". This excerpt from that quote is not specifically talking about a twist serve -- it is talking about any serve with slice.

You should be able to see the effect of the mild spiral spin on a regular topspin-slice serve. The slice (sidespin) component causes the ball to curve to the left (for a righty) prior to the bounce. However, quite often, we notice that the ball trajectory straightens out somewhat when the ball bounces. This is a mild shift to the right, since the ball is not going to the left as much as it was prior to the bounce. It is the ground interaction of the mild spiral spin that is responsible for this straightening out of the trajectory.

A Twist serve has a much stronger spiral component than either a topspin-slice or a topspin-kick serve. This is why the right shift on the bounce is so noticeable on the Twist serve.

For Twist serves that I have observed, the ball does curve somewhat from right to the left. This is due to the sidespin component. True, it may not curve in this direction as much as a typical topspin-slice serve, but it still exhibits a sidespin curve.

For more fun with physics: Rodney Cross home page

.