The fast low bouncing version of the Bigservesofthands first serve

[bigservesofthands]

Do you really play tennis?
You can just borrow or buy a soft racket, may I suggest the MGRadMid, at 56 flex? Hit with it for a week, then try a Pure Aero. Which is more accurate?

I have not played with many flexible racquets. Pure drive Roddick Plus, RF 98 autograph, Juice 100s and now Burn 100s. All pretty stiff racquets. I will demo some flexible racquets and see what they feel like.

I am not really interested in a flexible head and my point is not about the head. I would want the head to be as stiff as it is now. I am talking about making the beam directionally flexible. As an exaggerated case consider a nunchuck or a bull whip (but not as long) with racquet head attached to the end instead of a nunchuck handle or a bull whip end. The amount of head speed you would be able to generate with this setup would be so much more than a regular stiff beamed racquet. Now constrain the bull whip or the nunchuck chain from moving in planes that cause instability but allow is to shear/strain in ways that do not, and use that shear/strain release to increase your racquet head speed is what I am suggesting. There are materials (composites, graphene ...) available that are stiff in some directions and flexible in others that make such beams possible.

 

[LeeD]

I have not played with many flexible racquets. Pure drive Roddick Plus, RF 98 autograph, Juice 100s and now Burn 100s. All pretty stiff racquets. I will demo some flexible racquets and see what they feel like.

I am not really interested in a flexible head and my point is not about the head. I would want the head to be as stiff as it is now. I am talking about making the beam directionally flexible. As an exaggerated case consider a nunchuck or a bull whip (but not as long) with racquet head attached to the end instead of a nunchuck handle or a bull whip end. The amount of head speed you would be able to generate with this setup would be so much more than a regular stiff beamed racquet. Now constrain the bull whip or the nunchuck chain from moving in planes that cause instability but allow is to shear/strain in ways that do not, and use that shear/strain release to increase your racquet head speed is what I am suggesting. There are materials (composites, graphene ...) available that are stiff in some directions and flexible in others that make such beams possible.

Look for a Head Liquid Metal Radical Mid Plus. They're orange, tons around, probably sold 50,000 in the US. Very stiff hoop, flexible shaft, making a total of around 62 in flex rating.

 

[bigservesofthands]

Look for a Head Liquid Metal Radical Mid Plus. They're orange, tons around, probably sold 50,000 in the US. Very stiff hoop, flexible shaft, making a total of around 62 in flex rating.

Does not seem to be a current racquet. Unavailable except ****. I can only demo newer racquets.

 

[Raul_SJ]

I can serve in the low teens with this motion. Can crank it up to 120s at the cost of accuracy (mostly hitting long, directional accuracy is still present).

I think it would be a good option as a change up for a taller player who has played a pitching sport before and is in need of a fast easy action resting serve that does not require the more elaborate energy sapping service motion of a traditional serve

Are you able to throw a tennis ball from baseline over the fence? I have been practicing this a bit as throwing distance is supposed to correlate closely with serve speed. I have improved my throwing distance a few feet and can hit the bottom of the fence and my serve is stuck at 90mph. But I don't enjoy throwing more than 20 balls. It gets tiresome. But I might continue throwing if it is worthwhile and can help me get over 100mph.

 

[tennis_ocd]

Do you really play tennis?
You can just borrow or buy a soft racket, may I suggest the MGRadMid, at 56 flex? Hit with it for a week, then try a Pure Aero. Which is more accurate?

From what I read a highly experienced player will find the flex more accurate. Power comes with stiffness. But then strings provide a mag more "flex" than one will ever get with a frame.

Yet beginners find stiff and control go together.... A weird topic; I can't really note a difference but my elbow does.

 

[LeeD]

Good point.
I'm a low 4.0, so neither here nor there.
My racket's with flex of 56, 60 get 47 lbs. STB16,
My racket's with flex of 68, 70, 72, get 35 lbs. STB16, for some feel, dwell time, and easy swinging power.
I'm sure, if I strung my soft racket's at 35, it will be a mush sprayer. And if I strung my stiff racket's at 47, it would be an arm and shoulder breaker.

 

[bigservesofthands]

Since there was some interest on how spin serves look like using my wrist laid back style of serving, I had an opportunity to get them filmed. This was after a two hour playing session, when I drove to my in-laws who live four hours away to pick up my daughter who was visiting them. I had to drive back the same day (Sunday) so I did not miss work the next day, so did not have much time to get a lot of footage. It was a quick 5 min filming session. The receiver is my mother-in-law who got me into playing tennis, three years ago. She is a senior 4.0 player. Filming was done by my father in law who is a senior 4.0 player as well. Since I have a low memory iPhone 6 plus, I find having a human film is best to conserve space and get decent footage. It would be nice if I could find somebody to film me, in exchange for me filming them in Portland. Contact me if you are interested. So far I have not had much luck.

The first serve in the sequence is a low net clearance slice serve, the rest are all high net clearance top spin serves hit to body and down the T on the deuce side and body and out wide on the ad side. I usually am able to pull them in, into the box more, but I was a bit tired here after a long drive and a pretty hard two plus hour tennis session.

I am able to perform the traditional more vertical swing path serves with edge on to contact as well. These are faster on flats and kick up more on second serves. However these require a lot of vertical leg drive for me to get the ESR stretch needed to hit powerful serves. I am able to sustain this vertical leg drive only for around an hour or so. After that my legs refuse to drive and after a point I do not have enough time to drop the racquet still make it to the ball and strike without landing in the court. So I fall back on this horizontal swing path wrist laid back serves more and more as I tire and save the full on vertical drive serve for crucial points.

 

[Chas Tennis]

I looked at the serves and your racket motions for getting the spins. I did not see the strings moving the way I expected.

Here is a kick serve with the racket in a

1) frame before impact,
2) impact
3) frame after impact

Toly gif of composite video pictures from 240 fps video. The racket strings are rising.

Same for slice serve. The racket strings go mostly to the side.

Compare the string motions to yours.

Notice the forearm to racket angles.

 

[bigservesofthands]

I looked at the serves and your racket motions for getting the spins. I did not see the strings moving the way I expected.

Here is a kick serve with the racket in a

1) frame before impact,
2) impact
3) frame after impact

Toly gif of composite video pictures from 240 fps video. The racket strings are rising.

Same for slice serve. The racket strings go mostly to the side.

Compare the string motions to yours.

Notice the forearm to racket angles.

The traditional motions are obviously physically not possible with the wrist laid back approach due to the swing path and human anotomy. I am able to serve traditional kicks and slices too. However since this thread deals with an unconventional style, I posted how I am able to achieve kick and slice effect (though not as pronounced as traditional methods) using this unconventional swing path. It is physically much less taxing for me to serve these unconventional serves, so I continue to do so when I am tired and as a change up.

Here is a zoom of the contact on the second serve in the sequence posted above. It is a chest/shoulder level kick to a receiver standing 10 feet behind the baseline.

Watch it at at least 0.25 speed (gear icon in the youtube video bar in the bottom) to discern the amount of spin on the ball. A kick serve usually has about 4000 rpm. For a 240 fps video it is about a quarter revolution per frame. I estimate there is easily that much. Not sure how much is side, gyro and top. There has to be a decent amount of gyro+top since it drops into the box from a fairly high net clearance. The ball is also very heavy to the returner as can be discerned from how much the receiver's racquet shudders and then loses control of the ball.

Since the video is only 720p it is not possible to figure out the exact axis of rotation but the direction seems somewhat discernible. The ball is a "Penn 1" with only one "Penn 1" mark on it.

The contact itself is very complex and the ball pockets and then rolls off the string bed from the tip of the frame.

What I try to do to achieve this drop into the box with high net clearance and then kick effect is to time my ISR with contact and try to get the contact as much on the top left side of the ball as I can, after allowing the ball to drop enough to make that possible, but not so much that the serve goes into the net. A flat is a straight on contact attempt and the slice is top right contact attempt all while timing the max ISR to contact.

What happens in practice is as you can see, pretty complex. But this works with a high percentage for me and these are still a pretty offensive serves for most returners so I still start of the point at an advantage.

 

[Chas Tennis]

..................
Here is a zoom of the contact on the second serve in the sequence posted above. It is a chest/shoulder level kick to a receiver standing 10 feet behind the baseline.

................................
The contact itself is very complex and the ball pockets and then rolls off the string bed from the tip of the frame.
......................

I can't directly see the markings on the ball well enough to see the spin. My very uncertain guess would be underspin from one smudge that moves down the diameter in 4 frames, seen once. That would be about 1/2 ball rotation in 13 milliseconds or 2300 rpm. Underspin is also what I'd expect from the racket string motion on the ball. Needs quality video with very fast shutter speed where the ball markings are visible. The balls could also be marked better to show spin. Camera and other Youtube compression processing may affect seeing ball markings. When you view the camera video directly on your computer, without the additional Youtube compression processing, you will see the ball markings better.

Rod Cross, in his TW article on the kick serve (see "Improve" above on the TW webpage), says that the kick up comes mostly from the larger angle of the impact on the court. The ball bounces often shorter in the service box I believe. Your ball hits on the service line so it is less likely to have a very vertical kick bounce.

I believe the ball most always contacts the strings for only one frame at 240 fps so there would be no "rolls off the string bed from the tip of the frame". Video that from the side to see ball separation.

Your camera has a very fast shutter and shows the body and racket with small motion blur. But the unknown shutter speed may make seeing the ball markings difficult. ? The wide angle fixed lens on smartphone cameras often does not allow bigger images. Can the person shooting your serve videos stand closer without bothering your serve?

 

[Chas Tennis]

........................What I try to do to achieve this drop into the box with high net clearance and then kick effect is to time my ISR with contact and try to get the contact as much on the top left side of the ball as I can, after allowing the ball to drop enough to make that possible, but not so much that the serve goes into the net. A flat is a straight on contact attempt and the slice is top right contact attempt all while timing the max ISR to contact.
...................................................

This appears to be impact.

Where on the ball the racket first touches depends only on the racket face orientation, a plane meeting a sphere. As far as I can tell in this frame, the ISR that you mention has not started to rotate the racket to impact on the left side of the ball. Racket looks about vertical so contact might be slightly on the top half of the ball at most. ? Should be confirmed with a side camera view for more accuracy.

This frame is 7 frames later, ISR and pronation have rotated the wrist this far. The racket head appears to rotate after impact. This is only about 24 milliseconds after impact.

 

[bigservesofthands]

Here are zoomed in versions of nine frames. The first is right before contact, the next seven during contact (IMO) and the eighth is when the ball accelerates outwards after contact.

I can clearly see contact with racquet and ball pocketing and distortion of main strings as the ball slides from initial contact in the bottom middle of the racquet to when it flies outwards from the tip of the racquet. I can clearly see considerable ISR in the frames (shown here and the ones the have studied in the non-zoomed context) right before, during and after contact.

 

[Chas Tennis]

I can't see the strings in post #112. and believe that the impact lasts about 4 milliseconds.

Seeing the path of the ball is much clearer. I applied the red crosses manually for several frames using Kinovea.

Kick serve of posts #109 & 112.

Impact forces were applied to the ball very close to where the two lines intersect.

Here is also the first serve shown in the post #109 video, a slice serve. The video pauses for key images for 2 seconds. The video in between is played back as for the 240 fps recording.

Slice serve.

In the 80's when I was practicing serve, on very rare occasions the racket felt as if it contacted the ball for an unusually long time. Serve speed felt higher. Something rare was happening then..

 

[bigservesofthands]

I have uploaded the zoomed in ball contact videos for serves 3 to 5 in the sequence as well.

Zoom in of down the t attempt

Zoom in of ad body attempt

Zoom in of ad wide attempt

For all the wide kick serve attempts (serves 2, 4, 5) I can see significant dwell time of the ball on the string bed. 3-4 frames and the ball slides of the string bed from the tip of the racquet. I see significant gyro spin. It is more discernible for zoom in of serves 4 and 5.

So I don't believe that a long ball contact is unusually long for my out wide serve attempts where I really try to reach around the side the ball while the shoulder and the racquet are internally rotating.

Yes, for down the T attempt and for the slice the contact is only for around a frame but even that produces decent amount of spin but not as much as the out wide attempts.

 

[Chas Tennis]

I have uploaded the zoomed in ball contact videos for serves 3 to 5 in the sequence as well.
..........................................................
Zoom in of ad wide attempt

For all the wide kick serve attempts (serves 2, 4, 5) I can see significant dwell time of the ball on the string bed. 3-4 frames and the ball slides of the string bed from the tip of the racquet. I see significant gyro spin. It is more discernible for zoom in of serves 4 and 5.

So I don't believe that a long ball contact is unusually long for my out wide serve attempts where I really try to reach around the side the ball while the shoulder and the racquet are internally rotating.

Yes, for down the T attempt and for the slice the contact is only for around a frame but even that produces decent amount of spin but not as much as the out wide attempts.

I don't see evidence that the ball was on the strings for more one frame. The way to see that better is to video from the side and even try to view parallel to the string bed.

Last of the three serves.

 

[bigservesofthands]

It is possible that the dwell time could be an optical illusion. Chances are high that it is, because of the angle at which those serves were filmed.

I will try to get some footage from the side, for out wide kick serves hit with that swing path.

Ideally a set of iPhones synchronized to the millisecond shooting at right angles would be what is needed. Unfortunately I do not have such a set up. Maybe iPhone 7 will have 1080p 240 fps and we can see string details.

 

[bigservesofthands]

An issue of filming contact for serves from the side is that even at 240 fps the motion blur due to the rapidly accelerating racquet head makes things pretty much impossible to figure out what is going on at contact. The same issue is present in the side view of the first serve that was shown in the OP. After digging through my footage (that I get in bits and pieces at the end of sessions when a fellow player sticks around to film a few serves), I found one with that swing path (I am trying to perfect my traditional serves at the moment and most of my current collection is centered around that) where the motion blur is tolerable. Not entirely sure where the serve was aimed at. Likely a second serve based on recollection.

For this serve the contact seems to be between 1 or 2 frames depending on what you consider contact.

 

[Chas Tennis]

At 240 fps each frame is taken 4.2 milliseconds apart. With video shutters, exposure to light occurs every 4.2 milliseconds for a duration of time equal to the shutter speed. Light allowing, I try to keep shutter speed at 1/10,000 second, or 0.1 millisecond. Therefore, the exposure time to light at 240 fps would only extend to 4.25 milliseconds. Shutter speed is then very small compared to the time between frames and motion blur is small for tennis strokes.

For for low lighting, indoor tennis courts, cameras with automatic exposure control will select slower shutter speeds and we never know what the shutter speed was for the video clip. If your smartphone is also taking 240 fps with 4.2 milliseconds between frames and uses a slow shutter speed of, say, 2 milliseconds, then the time of exposure might extend to 4.2 + 1= 5.2 milliseconds. I have noticed that some iPhones have very fast shutters in bright sunlight, small motion blur.

(For objects moving across or up and down in the frame and not toward or away from the camera, motion blur can be used to estimate shutter speed. The length of the motion blur relative to the length of object motion between frames, a percentage, gives a crude indication of the shutter speed for that video. If the motion blur is 50% of the distance that an object edge moves, then the shutter speed lasted 50% of the time between frames in those frames.)

Ball to string contact lasts probably 3-5 milliseconds.

If your shutter speed is slow it will increase the chances of capturing the ball in two frames. But the motion blur from the side view increases and would be bad for telling initial ball string contact and separation times, make those times very inaccurate. It is hard to tell initial contact or separation in a frame even with a shutter speed of 1/10,000 sec.

The best approach is to get a better video in direct sunlight and see if the motion blur is workable to see what you want. That way your camera's automatic exposure control will select its fastest shutter speed. You still have to interpret the sunlight motion blurs, and they are very uncertain for ball-string contact. On the other hand, if the ball blur is clearly away from the racket face then you know that, but just not very accurately. You will probably be able to see separation.


[Why take video in bright sunlight and not indoors? There's 100X as many photons in the sun light! -
http://www.kinovea.org/en/forum/viewtopic.php?pid=2033#p2033]

 

[bigservesofthands]

I was searching for the current state of the art in high speed capture that is somewhat accessible to amateurs. Lo and behold! I came across this

It answers the question we were discussing. After initial contact the ball clearly rolls off pronating string bed that is moving downwards following the trajectory of the ball and shoots off the tip of the racquet, like I was saying, for a number of serves shown in that clip. Contact lasts for at least 4-5 frames at 240fps for the serve at 0:08, at least 5 frames for the 240fps serve at 0:10, at least 14 of the visible 960 fps frames, while the racquet goes out of the video frame and the serve contact is not even a third done yet.

Here is an animated gif of the contact of serve at 0:08. It skips 1 frame due to some quirk in Quicktime during frame by frame forward, but that frame is visible when you go backwards.

So yes the racquet can be moving down on contact (while rapidly internally rotating) and still generate gyro and top spin as one can see on the ball release on serve at 0:08

It is funny that not just that string bed, but the rim of the frame is also involved in a lot of spin serves. Maybe something gritty on the tip of the frame might help get a bit more spin!

This is a big discovery for me!

Here are my conclusions.

a. Saving maximum ISR for contact does help.

b. Approaching the ball with the wrist laid back is quite alright to generate a high level serve, as long as you have enough ESR stored up to release on contact. It validates to me that my horizontal swing path technique is a sound approach. It explains empirically the results I see in practice with this approach.

c. The racquet behaves as a large open face extension of your palm and you are able to perform a throwing motion using it imparting various spins your palm is capable while your shoulder/forearm/wrist rotate rapidly.

d. A serve in a lot of cases is a controlled push of the ball using a racquet that is moving at a high RHS. It need not be a straight on contact. It is likely that most high level serves including first serves are controlled pushes and you need to avoid full on contact if you can, if you want a controlled high margin serve.

e. You don't really need to hit up of the ball. What you need to have enough ESR stored to push the ball forcefully forward while brushing around it rapidly, while the balls rolls across the string bed and enough forward RHS to keep the accelerating ball on the string bed as long as possible.

I think I will start saving up the 1500 bucks for the Sony RX10 III. Seems like a nice camera. Or maybe others like Canon or iPhones will catch up soon.

 

[bigservesofthands]

Here is the animated gif for contact of 240 fps serve at 0:10

It shows the sliding ball pretty clearly.

I use the racquet tip to aim and this validates the soundness of that approach as well.

 

[bigservesofthands]

Here is the 960 fps contact that clearly shows the ball being pushed and sliding across the rapidly rotating string bed.

 

[Chas Tennis]

The shadow of the ball is visible in the videos and in your gifs in #120 & 121.

Red arrow indicates ball shadow. Blue lines indicate reflection of ball from frame, less light in ball shadow gives dark area on frame. The ball shadow makes the strings darker against the white background, with the improved contrast the strings become visible in the ball shadow.

When the ball is near or in contact with the strings the ball and shadow touch. When the ball and shadow are separated the ball is separated from the strings.

Same as for first image, ball blocks direct sunlight from reaching an area of the strings and the frame (indicated by the blue lines). Look at your gifs to see the shadow move.

The SONY RX10III is capable of some great videos. One issue I have found with some recent SONY models is that the high speed video recording time is limited to 2 or 4 seconds. Check and consider that, it's a show stopper for me. Maybe if you had a camera operator it could be workable for tennis strokes, but I doubt it. In the SONY Youtube, the camera was hand held as you can see by the background moving.

The basic problem is that the camera gives a 2D image of 3D space. The components of motion are well shown up-down and across the frame. But the 3rd dimension, toward or away from the camera, is reduced sometimes very strongly. If you want to see ball separation from the strings, set up the camera so that separation is across the frame.

If you view from the side in bright sunlight using iPhone and moving the camera you probably will be able to see your separation with 240 fps. You want to view parallel to the face of the racket as best you can.

I've seen only one serve where the ball and racket had much closer speeds. The ball and racket seemed to travel at nearly the same speed for many frames at 240 fps. I have wondered if that was a busted ball. My rough estimate is racket at 100 MPH for ball at 125. I believe that Yandell had an estimate similar to that. It could be measured in percent by looking at a few flat serve frames.

I believe that I felt some very rare longer contact times in the 1980s probably with my Waiter's Tray. I assume that timing of racket acceleration was the reason. ? Maybe you sometimes get that and can feel it. ?

The Casio FH100 (last offered new 2011) has done everything that I have wanted, It records for 11 minutes at 240 fps and has a manual shutter to 1/40,000 sec.

Casio has a new camera, Ex 100f, the only affordable camera to offer high speed video with manual shutter speed control since the FH100. Shutter speed to 1/10,000 sec. ~$600

It does not have as high resolution at 1000 fps as the SONY.
http://www.kinovea.org/en/forum/viewtopic.php?pid=3598#p3598

 

[Chas Tennis]

duplicate post deleted

 

[bigservesofthands]

Yes after looking at the ball shadow that has been pointed out and on closer inspection by zooming in on contact, it appears that the 240 fps contacts in serves 1 and 2 last for little over a frame. There is no frame involved in the contact.

The contact (5 frames) in the 960 fps serve 3 shows much more detail and here one sees what I expected (albeit on smaller time and space scales), the ball rolls/slides across the string bed (six or more main strings) as it is pushed outward. The contacts is completed in the video and the ball flys outward with no frame involved.

The strings and string bed seem to be pretty stiff and tight. Not much ball pocketing or even ball distortion is evident.

The distances separating the ball from the string bed as the racquet face rotates across the outgoing ball seem very small. Fractions of an inch.

Given such small margins and assuming ball pocketing (clearly visible in my lower resolution serves) and faster angular rotation of the racket, I cannot rule out the possibility of longer dwell time and multiple contacts on the string bed when using my serving style and string bed setup. I can never be sure until there is better resolution footage.

 

[JohnYandell]

If you study high speed video it shows that the universal dwell time is right dead about 4 milliseconds. To see the pocketing you need to film at like 10,000 frames per second. There may be slight differences in dwell times but how that relates to swing speed, strings, incoming ball speed, incoming ball spin type of ball, weather, etc is unknown and probably unknowable. the interesting thing is how constant the dwell time is. The idea that you can somehow "hold" the ball on the racket or that you create more dwell time somehow and that effects the shot, that's a myth.

 

[JohnYandell]

The ball definitely does not roll. At 10,000 frames a second which is what you need to really see, the ball pockets deeply and embeds itself in the strings. The strings may displace like 1/4 inch or something, but the ball is going right with them.

 

[coupergear]

Lol. Awesome overhead whip action. Can't say I've ever seen a waiters tray quite like that one.

Sent from my SM-G930T using Tapatalk

 

[Chas Tennis]

Lol. Awesome overhead whip action. Can't say I've ever seen a waiters tray quite like that one.
..........................

That is what I thought, at first..........

Better definitions for Waiter's Tray technique are needed. My current definitions:

1) High Level Serve - technique that uses internal shoulder rotation (ISR) as a significant contributor for racket head speed. Among all high level servers between Trophy Position and impact there are similar motions with similar biomechanical functions. In addition, before Trophy Position most servers, but not all, also develop racket head speed that is considerable. High speed videos clearly show two significant racket head rotations at impact, one from ISR and the other from swinging.

2) Waiter's Tray - Lower performance technique that uses little internal shoulder rotation (ISR) for racket head speed. High speed videos will show one racket head rotation at impact. The rotation is not a simple single rotation, it is around several axes going through the wrist, elbow, shoulder, waist, etc.

3) Waiter Tray Like plus ISR Combination (bigservesofthands, Raul_SJ) - Lower performance technique that uses significant internal shoulder rotation for racket head speed. High speed videos will show mostly one racket head rotation at impact similar to the Waiter's Tray, face up then to face on at impact.

High speed videos show both 2) and 3).

The most important stroke technique issue in tennis today is the high level serve vs Waiter's Tray. The definitions above need to be improved to better understand the lower performance techniques in use and move things along. The serves that are being used with statistics are needed. What % use 2) & 3) above?

Of course, since the Waiter's Tray is not properly defined anywhere, there will always be endless discussions when arguing undefined terms. Is the WT definition based on the racket head orientation, face up to face on at impact, or the absence of significant ISR?

In my view. the current high level serve technique of top ATP & WTA servers is well defined biomechanically by high speed videos from the Trophy Position to impact.

 

[bigservesofthands]

deleted, duplicate post

 

[bigservesofthands]

deleted, duplicate post

 

[bigservesofthands]

The ball definitely does not roll. At 10,000 frames a second which is what you need to really see, the ball pockets deeply and embeds itself in the strings. The strings may displace like 1/4 inch or something, but the ball is going right with them.

Thanks for taking the time to respond. You have devoted far more time and resources to the problem and have much more data at your disposal. So your word counts higher than mine. You have data to disprove my assertions and I don’t.

When I serve with extreme ISR close to contact, the feeling I get when I pay attention to the contact is of the ball rolling on the strings, pocketing and rolling out in the direction I am trying to get the ball to go.

This is borne out in the open source high speed video of tennis serves that I have found.

Here is the zoomed in 1000 fps contact for what appears to be a kick serve based on swing path shot using the Sony rx 10 iii that I posted above

The cameras that professional sports science researchers are currently using seem to be the Olympus i-speed series 7 cameras. The video that Toly posted for the 140 mph serve contact seems to be from this video from the Olympus promotional material

Here is a nice 1080p 7500 fps video of a talented young man's serve using a wet tennis ball that was posted recently which was shot using the Olympus i-speed series 7

One can clearly see the rolling in from lower left, pocketing a little to right of center and rolling out from lower left fairly close to the frame. Here is the zoom in contact

Toly posted a video of a much more drastic rolling and double bounce on a string bed in the past

So it is not out of the realm on possibility

 

[bigservesofthands]

I have come to the realization that the optimization path for the wrist laid back, horizontal swing path and mostly horizontal delivery while facing and aiming at the target style of tennis serve, lies more with David's slingshot and buggy whip techniques.

When hurling a rock using a David's sling shot or using a whip to hit pin pin point targets fine directional control with high speed delivery is achieved. You need the same in a tennis serve. So you can see this evolution in my wrist laid back serves.

I continue to pursue traditional serves as well.

 

[Chas Tennis]

delete duplicate post

SKIP TO POST #141 _ ERRORS POSTING_DUPLICATE POSTS DELETED

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

duplicate post -deleted

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

duplicate post - deleted

 

[Chas Tennis]

Thanks for taking the time to respond. You have devoted far more time and resources to the problem and have much more data at your disposal. So your word counts higher than mine. You have data to disprove my assertions and I don’t.

When I serve with extreme ISR close to contact, the feeling I get when I pay attention to the contact is of the ball rolling on the strings, pocketing and rolling out in the direction I am trying to get the ball to go.

This is borne out in the open source high speed video of tennis serves that I have found.

Here is the zoomed in 1000 fps contact for what appears to be a kick serve based on swing path shot using the Sony rx 10 iii that I posted above

The cameras that professional sports science researchers are currently using seem to be the Olympus i-speed series 7 cameras. The video that Toly posted for the 140 mph serve contact seems to be from this video from the Olympus promotional material

Here is a nice 1080p 7500 fps video of a talented young man's serve using a wet tennis ball that was posted recently which was shot using the Olympus i-speed series 7

One can clearly see the rolling in from lower left, pocketing a little to right of center and rolling out from lower left fairly close to the frame. Here is the zoom in contact

Toly posted a video of a much more drastic rolling and double bounce on a string bed in the past

So it is not out of the realm on possibility

Toly has an interest in the unusual. He often posted exceptions.

I'm overwhelmed in studying what is happening with the top servers 99+% of the time. I try to avoid tricky and ambiguous interpretations of high speed videos especially the ones where I can't find reliable descriptions by the people that produced the videos.

1) A wet tennis ball I'd assume has different friction and could not be assumed to represent a dry ball impact. Also the water spray makes contact difficult to see. How do you know the ball was on the strings for more than 4 milliseconds or moved? Post the two frames that unambiguously show separation or rolling.

2) Toly found a double hit where the racket does not appear to be moving relative to the camera. What were the circumstances?

If you have a serve where the ball stays on the racket strings longer the only way to make a creditable case is to show some clear videos of your serves. Those would probably be side camera views with very high frame rates and extremely fast shutter speeds. That way the direction of contact and separation, forward and backward would be viewed across the camera frame and be easier to see than toward the camera.

There are many very high speed videos on ball string contact in the TW information above on this page. They show strings stretching and moving to the side. Example,
http://twu.tennis-warehouse.com/learning_center/stringmovementPart2.php

Rod Cross has worked in this area and probably has publications. Use Google Scholar to search. Include Researchgate to find and get publications.

 

[JohnYandell]

Yeah Pete is an old friend of mine and I also know Rod Groom. In that first gif at 1000 frames the ball is on the strings probably only the last 3 frames. I posted some 10 or 12 thousand frame video of the pocketing but I cannot post from my own site now. It was way back when maybe someone can find it.
When the ball pockets the strings distend--this is the cause of the poly snap back and this causes the whole pocket to "move" slightly--but maybe half the gap between two strings--it's the whole pocket that is sliding--you can see this deep embedding in one of the videos below the ball itself is definitely not turning over or sliding across the surface.

 

[bigservesofthands]

Yeah Pete is an old friend of mine and I also know Rod Groom. In that first gif at 1000 frames the ball is on the strings probably only the last 3 frames. I posted some 10 or 12 thousand frame video of the pocketing but I cannot post from my own site now. It was way back when maybe someone can find it.
When the ball pockets the strings distend--this is the cause of the poly snap back and this causes the whole pocket to "move" slightly--but maybe half the gap between two strings--it's the whole pocket that is sliding--you can see this deep embedding in one of the videos below the ball itself is definitely not turning over or sliding across the surface.

My claim for extra dwell time is for spin serves especially those that are hit out wide on both courts either through slice/top-slice/twist

Here is what I see for the Sony rx10 1000 fps serve:

It is a top slice serve. First frame is before contact. Second frame is contact albeit barely (which can be inferred from the fact that the shadow of the ball is partial and intersects the ball from our viewpoint). The third frame is beginning of roll, the ball had no spin before that and from frame two to three it has rolled maybe 5 degrees on the strings. The fourth frame is continuation of roll and between frames three and four the ball has rolled a further 8-10 degrees. The server has initiated ulnar deviation between frames four and five and the racquet head suddenly accelerates from it. When I serve traditional top slice serves, I go by both sight and feel and can feel the ball on the strings which is my trigger to initiate ulnar deviation. The ball is pushed by this ulnar deviation while getting embedded significantly more into the bed by frame five. Between frame five and six the pocket itself moves by at least main string and the ball rotates in the pocket and comes off the pocket somewhat. The ball is clearly off the string by frame 7 and if we guesstimate 2 inches of clearance between frame 6 and 7 then we get a impact serve speed of 113 mph. Seems a little less than 2 inches so maybe a 100 mph or so serve.

Here is what I see for the Olympus i-speed 7 7500 fps serve with the wet ball. Likely flat serve given the much higher ISR brought to bear and lack of any discrete ulnar deviation.

After the first no contact frame, I can clearly see slight contact, then a roll for a frame then ball pocketing with maximum impact pocketing in the lower middle of the frame and then the pocket moving to the bottom left and the pocket pushing out and finally releasing the ball only in the tenth frame after contact and that too towards the bottom left fairly close to the frame.

Even higher resolution and higher frame rate images are needed to be conclusive. A wet ball contact might change the properties.

 

[JohnYandell]

Yeah I just don't agree. It sounds logical but I don't see it in our high speed footage. But we can agree to disagree. We are talking about truly invisible events.

The serve contact is sometimes a little shorter if anything than the groundstrokes. Maybe a little less than 4 milliseconds, so even less time to "slide". At a thousand frames a sec if that's what that clip is the last 3 frames are the only ones with contact. You can tell by the slight change in the shape of the ball. At 10 times faster you would see how the ball embeds deeply. I just looked at a Sam Stosur kick serve that was filmed at about 7500fps. The ball rotates slightly in the "pocket" so the movement from right to left when viewed from behind is about 3/8 of an inch I am guessing, about the distance between the strings. The racket is moving radically left to right and slightly upward to generate the spin. This is I think causes the confusion--the racket is moving directionally at contact and that creates the illusion of a lot of roll.

I will have some of these clips made into free pages when our guy that does that gets back in a week or so if I can email them to someone else to post. We can't put them on YouTube due to certain copyright restrictions, but I can create a link.

 

[JohnYandell]

PS: take the Pete McCraw video and drop it into rowvid and view it frame by frame.

 

[bigservesofthands]

Yeah I just don't agree. It sounds logical but I don't see it in our high speed footage. But we can agree to disagree. We are talking about truly invisible events.

The serve contact is sometimes a little shorter if anything than the groundstrokes. Maybe a little less than 4 milliseconds, so even less time to "slide". At a thousand frames a sec if that's what that clip is the last 3 frames are the only ones with contact. You can tell by the slight change in the shape of the ball. At 10 times faster you would see how the ball embeds deeply. I just looked at a Sam Stosur kick serve that was filmed at about 7500fps. The ball rotates slightly in the "pocket" so the movement from right to left when viewed from behind is about 3/8 of an inch I am guessing, about the distance between the strings. The racket is moving radically left to right and slightly upward to generate the spin. This is I think causes the confusion--the racket is moving directionally at contact and that creates the illusion of a lot of roll.

I will have some of these clips made into free pages when our guy that does that gets back in a week or so if I can email them to someone else to post. We can't put them on YouTube due to certain copyright restrictions, but I can create a link.

Thanks for offering to put up some of the clips into the free pages. That would be useful to people like me.

Is this the video of the Stosur Kick Serve contact that you are talking about?

If it is then thanks for clarifying the frame rate. Toly never mentioned it as far as I know.

Here is the zoom in animated gif of contact

I see a slightly bigger amount (an inch and a half to two) of roll than the distance between strings (i am counting the difference between the farthest point at release to farthest point at contact). But yes the the contact is much quicker in this serve if the frame rate is indeed 7500 fps. A quick guess estimate is an 80 mph, 3000 rpm serve (using the Wilson ball markings as a reference since they are of a known size).

Not sure what the target of the serve was. I experience much shorter dwell time on down the T twist serves versus out-wide where you are trying to reach around the ball more.

My conjecture is if one can bring to bear even faster racquet rotation rates to bear than Stosur is, then the dwell times, spin rate and exit velocity would be higher than what she is able to obtain. She seems to have a continuos racquet swing at contact. There is no wrist release timed closed to or at contact.

It would be good to see contact videos for more serves, especially men (Isner, Thiem, Federer ...). I am guessing in the next five years this sort of power would be in the hands of amateurs in consumer grade devices that they use regularly so such footage might become common place. Even quicker if enthusiasts and instructors put it up sooner.

 

[bigservesofthands]

PS: take the Pete McCraw video and drop it into rowvid and view it frame by frame.

The one in that video is a straight on flat serve aimed dead center on the ball. Slightly down on contact not much spin, not even gyro. Likely hit the dead spot and very good momentum transfer as a result. Look how much the racquet head slows down after contact.

Yes slightly higher dwell time than Stosur kick because of deeper and longer embedding and not much roll.

This is the feeling I get on my flats as well.

 

[JohnYandell]

I think we can just agree to disagree now. That's not the clip I have but on this Stosur clip the ball isn't rolling. Racket is moving. There is no such thing as a flat serve with the correct motion. The lower clip is a lot more frames and that is actually what happens to the ball on all serves. That pocketing. Notice the racket head is rotating. Look at the label on the front of the ball. It disappears to the left and the 8 rotates into view. That clip might be 10000 frames at that rate one rotation is going to look really slow.

 

[mbm0912]

I think about everything in this thread before each ball toss. It's really helped.

 

[Digital Atheist]

^^ Then I'm betting you break the 20/25s time violation rule!