Racket angle at contact

[marius hancu]

I know that you have high-speed cameras, thus I am pretty sure you have frames showing the racket right at contact.

Somehow, this is rarely shown in the clips at this or other sites (talking mainly about .mov clips). Normally what we can see by single-stepping or slow-mo in Quick time is the racket at 1 foot from the ball and it at another foot after the contact:-)

Now placing a full-stroke in a clip filmed at high speed on the server would very probably require a large size for the clip and thus download probs, etc.

But perhaps you could take a short slice from Federer's inside out FH, just around the contact, and present it in a clip. I would prefer of course a lateral view, in order to see the angle of the racket (if any) from the vertical. Same for other shots (rally slices, etc).

[johnyandell]

High Speed Clips

You are 100% right. Statistically you have about a 1 in 8 chance of getting the ball on the racket with the 30 frame video. That's one of the big advantages of the high speed DVDs--and we will, over time be using more of that footage on the site.

Now the issue of the racket face at contact is very tricky. I've seen some of those still pics in the TW message boards and in most cases they definitely represent mishits and this causes the angle to deflect, and people think "wow, so and so hits with a really closed face--I'll try that..." Disaster.

From what I can see there is no absolute right or wrong racket face angle at contact. In general the old school thinking that the face is vertical at contact is proved out in the great majority of the groundstrokes in the high speed footage--but not always.

There are high balls where the face is clearly closed up to 15 or 20 degrees moving to the ball and actually at contact.
For the average player though this may be totally irrelevant. (For a more aggressive view on the uses of racket tilt, see Kerry Mitchell's articles on the grips.)

My friend Rod Cross the Australian physicist sent me a paper describing the relationship between racket tilt, ball speed and spin. Unfortunately it was written in equations. But if the ball speed, height, and swing plane are just right, the closed face is apparently physically possible or even correct. I've stayed away from addressing this issue until I understand it all better. But it would be a mistake based on the high speed footage to say this was the norm or something to model--even if you were a tour pro, much less for the rest of us.

[marius hancu]

But if the ball speed, height, and swing plane are just right, the closed face is apparently physically possible or even correct. I've stayed away from addressing this issue until I understand it all better. But it would be a mistake based on the high speed footage to say this was the norm or something to model--even if you were a tour pro, much less for the rest of us.

Well, I think all of us need to understand this better and my own experience and knowledge a physics (I'm not a physicist, but took enough courses in physics) leads me to believe that indeed one can slightly close the racket face depending on some other physical parameters of the shot.

Now, question is, do you expect to be able to create for this site several "zoomed clips in the contact area?" running under QuickTime? They would be very useful even in the raw state, as experimental results.

[johnyandell]

Zoomed

Not quite like that but something that shows what you're talking about.

[vmiller]

You guys may already know this, but I think you can read more about this topic in 'The Physics and Technology of Tennis' by Howard Brody.

There's such a small window in which the trajectory of the serve must come off the racquet to land in the box (somewhere in the neighborhood of 8 degrees I think), I can't imagine trying to dictate the angle of the racquet at contact, especially considering the speed of the swing. Lucky for us that our bodies are able to figure it out. I guess it would be something interesting to see though.

[marius hancu]

Well, based on this high-speed camera clip from the Advanced Tennis series at the site (suggest running the clip in slow-motion at the original page, if you can't do it here in the forum), I think we can conclude that:

- the racket face is NOT vertical at contact; the angle with the vertical is about 10-15 degrees, I guess;

- the ball does not hit the racket right in the middle of the sweetspot (intentional design by Federer? very probably so, by training), but in the upper half of the racket (i.e. the highest wrt to the ground in this situation; mind you, the racket axis is somewhat horizontal at that point in time)

- during the contact with the racket (4 milliseconds or so), the ball travels towards the lower (wrt its axis) half the the racket; this is where an important contribution to the spin of the ball is made; this is an assertion which I'm a little bit less confident of, based on the clip which is not clear enough in this very respect

Of course, I'd be reluctant to generalize to other FHs that Federer hits or to other players, but this is one of his typical ones.

[johnyandell]

Great job with the post Marius!!

This clip shows that it is more than possible to hit with the racket head "tilted." In fact, Rod Cross the noted Australian physicist is doing an article for us on the physics of this, and the relationship between ball speed, spin, racket tilt, and shot trajectory.

Based on lookiing at a few dozen other high speed forehands by Federer, and a few hundred in 30 frame video, (and hundreds of high speed strokes of other players who occassionally do the same) I'd say it's a variation rather than the norm--but certainly not a fluke or exception. The question would be why this one and what is different, but these are things I'm not prepared to really go out on a limb about, just yet.

This one isn't a mishit. You are seeing the frame right before and right after the hit, not the actual hit. This is due to a shortcoming in how the 250 frame video renders when not viewed on the source tape with a special deck. It looks like the ball hits above center. The reality is that when this happens you see twisting in the frame--sometimes quite violent, up to about 90 degrees. In this case the angle of the frame stays constant so he got it in the middle. You can extrapolate this imagining the ball path.

We would need to film much faster 2000 frames or something even higher to really see how the ball moves on the strings. My impression is not much. Other high speed examples in the archive catch the frame on the strings--the ball seems to come directly off from the contact point, but it could be moving slightly with the change in spin--maybe it has to? Rod may be able to answer that.

[marius hancu]

This one isn't a mishit. You are seeing the frame right before and right after the hit, not the actual hit. This is due to a shortcoming in how the 250 frame video renders when not viewed on the source tape with a special deck.

Well, I do what I can without that special equipment which I am sure is necessary:-[[

[johnyandell]

Many many mysteries still to investigate...you mean you don't have a 10,000frame/sec camera????

[michaelm704]

Closed face? He's hitting on the rise...

Folks, physics person here, but you don't have to be one to notice that Roger's hitting the ball on the rise. With the ball on the way up - if the face was vertical, the shot would probably have sailed long.

[CrazyLefty]

Probably out

The path of the racquet head is also rising with a greater velocity and weight than the ball. The racquet face could be vertical at contact and still impart topspin on the ball, keeping it in the court.

[johnyandell]

Yeah not sure the on the rise is the key, as lefty points out. Although it is not the usual racket head angle it does happen fairly frequently in pro tennis on high balls especially. The noted Australian physicist Rod Cross is writing an article that will be up in the next 2-3 months that should give us all a much bett framework for understanding this. For years the vertical racket face has been a sacred cow--and I still think for most players on most all balls it's the way to go. BUT for a little different view, check out Kerry Mitchell's articles on the grips.
http://www.tennisplayer.net/members/...and_Grips.html

[CrazyLefty]

What about the racquet movement?

I do not agree that a rising ball hit with a vertical racquet face will probably go out. The racquet is moving much faster than the ball in the vertical direction. The racquet also has much more weight and is impacting the back of the ball. This would reverse the spin, of the ball, imparting topspin, and make it possible for the ball to stay in the court. It is possible for the face to be closed or vertical at ccntact and the movement of the racquet head prior to contact must be taken into consideration.

[gzhpcu]

We are also witnessing very fast forearm pronation. On the forehand, the racket face closes very quickly after impact. I would guess that it is impossible with all this forearm pronation to time a perfectly vertical angle for the face of the racket. Also, as John mentioned, the racket head will twist for off center impact. Also impossible to hit dead center all the time. IMHO, this discussion was relevant in the old days, when topspin was hit with no or very little forearm pronation, and primarily with the low to up movement, but not today for advanced tennis. If only we could observe the ball during the period it is squashed against the strings...

[johnyandell]

The face doesn't necessary close much or at all immediately after the hit. It can stay on edge for many frames after impact (in the high speed footage). Check it out for yourself in the high speed archive. (Safin, Agassi, Hewitt are the clearest examples.) This is true even with a fair amount of hand rotation. What I see in the high speed footage is that the face can go into the contact either vertical or slightly closed. (Federer and Justine are closed in the examples in the archive--but remember these are just one example each--and we can find the vertical face too.) It must be possible to deal with many balls with either version--the question we don't know wht the real difference is--yet. I would say that in general the vertical face is still overwhelming prevelant.