Roger Federer and the Evolution of
the Modern Forehand

By John Yandell

Torso Rotation

In our previous analysis, (Click Here) we saw a direct correlation between grip style and torso rotation. In general the more extreme the grip, the more torso rotation. A player like Sampras rotated about 90 degrees or a little more on the forward swing, with the shoulders normally finishing something like parallel to the baseline. Players like Agassi and Safin, with moderate semi-western grips, rotate further with the rear shoulder, coming around slightly past parallel to the baseline or slightly further. Players with more extreme semi-western grips like Hewitt, Guga and Roddick rotate further still. At times their shoulders rotate 180 degrees or more on the forward swing, sometimes finishing with the rear or right shoulder pointing at the opponent.

According to that gradient, we would expect Federer’s forehand to have a moderate overall rotational pattern, probably somewhere between Sampras and Agassi. And sometimes that’s true. But at other times, his rotational pattern looks like the most extreme players such as Roddick. Or it can be somewhere between.

So you’ll see him rotate about 90 degrees from the turn position to the hit, stopping with his shoulders about parallel to the baseline. Other times, the shoulders don’t stop and come all the way around, rotating 180 degrees or more with his back shoulder finishing facing his opponent. Or it could be halfway in between. Again all three rotational patterns can be paired with all the variations in the hitting arm position, and any degree of hand and arm rotation. So to say the least once again, Federer simply doesn’t fit within the parameters of our paradigm.

Interestingly the one constant seems to be his shoulder position at contact. This is consistently parallel to the baseline, or very close to it. This is yet another element that is different than we might predict based on his grip. We saw that Sampras was closed about 45 degrees or a little less at contact. Agassi was usually closed about 30 degrees. The more extreme players were open at contact, with their shoulders generally parallel to the baseline. In all these cases, whatever the angle of the shoulders at contact, the contact appeared to be about in the middle of the overall rotational pattern.

Federer doesn’t do that. His shoulders are consistently parallel to the baseline at impact, what we would have previously considered to “open” for his grip style. The position of the shoulders at the contact point doesn’t seem to vary much whether or not the shoulder rotation stops or continues after the hit. It doesn’t seem to vary with the different hitting arm positions either. He can extend the hitting arm in the straight position. Or stay vertical and hit the classic finish position, or snap it off with a sharp right to left rotation of the hand and racket head and finish on his left side. The shoulders stay parallel at contact or very close. The only possible exception is on some inside balls hit with the straight arm position and extreme total rotation. In those cases, his shoulders can sometimes appear to rotate somewhat further, so that they are somewhat past parallel to the baseline.

Hand and Arm Rotation

We’ve already referred to Federer’s hand and arm rotation quite a bit, but let’s look at it in more detail, and put it in the context of what we’ve seen with the other players. In our previous articles, we identified the rotation of the hand and arm as a key variable across the grip styles and found that the more extreme the grip, the more hand and arm rotation players used on their basic forehands. We also found that the players with extreme grips increased their hand and arm rotation further when they wanted to increase spin, hit angles, and deal with low balls or wide balls. The classical players, by comparison, had much less total hand and arm rotation. They also tended to vary it less when dealing with different types of balls.

We can see the amount of hand and arm rotation in a given shot by looking at the position and angle of the racket tip as it moves through the swing. Once the hand arm and racket reach the hitting position, they will rotate to the player’s left, or counter clockwise during the forward swing. To understand it, you may want to check the Advanced Tennis article that isolates just the rotation. Then it’s easier to see in the context of the swing. (Click Here.) For example, if the tip of the racket starts pointing directly to the back fence at the start of the forward swing, and finishes pointing directly to the sky, that’s 90 degrees of rotation, about standard on a classical forehand.

When there is more rotation, the racket tip won’t stop when it is pointing up, it will continue to rotate to the left. The more internal rotation, the further the tip goes to the left. At the same time the racket face is turning upside down. What this means is that the top edge of the racket at the contact starts to turn over and point back toward the player. In the extreme case, the racket turns over completely so the face in on edge to the court and the tip ends up pointing at the left sideline, or even slightly backwards toward the baseline.

It won’t surprise anyone at this point to hear that when it comes to hand and arm rotation, once again Federer has it both ways, or more accurately, many ways. It’s starting to sound like some kind of repetitive mantra but it’s also true: Sometimes it’s classic, sometimes it’s extreme, and sometimes it’s at various points in the middle. This is exactly same kind of flexibility and variety we saw in his hitting arm position and his shoulder rotation.

This rotation of the hand and arm is the reason so many observers immediately conclude that Federer has a semi-western or even a western grip. These so-called “windshield wiper” finishes can be as extreme, or even more extreme, than those we are used to seeing from players like Juan Carlos Ferrero, or Andy Roddick or Gustavo Kuerten. But the reality is that Federer is doing this with a much more conservative grip. With the extreme grips the hand and arm rotation is a basic component of almost every swing. With Federer it’s an option. He’s more than capable of hitting through the ball with a classic finish.

Since there has been so much confusion about what is actually happening with the hand, arm and racket rotation, and how it effects the finishes in Federer’s forehand, let’s try to understand it as accurately as possible.

Let’s start with the simplest case in which this hand and arm rotation is closer to the classical model. On some balls, Federer’s finish could be the model in a textbook on old school eastern tennis. He can finish with his arm in the classic double bend position, with the racket on edge, and the wrist at eye level. This finish indicates that there has been about 90 degrees of hand and arm rotation, which is associated the classical forehand. At this point his racket has reached the universal finish position that we identified in the article on commonalities across the grip styles. (Click Here.)

You see Federer make this finish on some balls in the center of the court. Also on short, relatively low balls. Sometimes on these balls, he also steps directly into the shot with a neutral stance. You can see the same finish on inside balls, especially if he is moving backwards or playing a more defensive ball. You also see it on his forehand returns. It probably accounts for 20% or less of the total finishes on his forehands.

But those aren’t the Federer finishes people notice. On most of his forehands, he incorporates significantly more hand and arm rotation. This is what catches the eye of fans, players, and television commentators, particularly the more extreme examples, where the racket turns over the most. When this happens the wrap at the end of the swing also takes the racket further across the body to his left side.

There is no doubt that with the extreme rotation, the wrap is definitely lower and further across the body. Sometimes, however, commentators and other observers describe this wrap position as much more extreme than it really is. This is the so-called “finish in the pocket” where he hand at the end of the wrap is supposedly at the same height as the pocket in his shorts. In all our filming, I’ve yet to see that. Occasionally, you’ll see him finish around the center of his torso, but that is as low as it gets. More typically the wrap finishes a little below shoulder level. From that point, Federer starts to recover, putting the left hand back on the racket and sliding it up the shaft to the throat as he moves back toward the ready position.

The left side wrap is characteristic of all forehands hit with more internal rotation of the hand and arm, whether by Federer or Roddick or whomever. But it’s critical to understand that the left wrap is actually the deceleration phase of the swing. This means that during the wrap the racket is actually going slower than at any point in the swing. This is where a lot of people who try to model Federer make a mistake. The left wrap may look dramatic, but it’s not the key to racket head acceleration. Its function isn’t to accelerate the racket, it’s is the opposite, to decelerate the racket. The left wrap is the result not the cause of racket head acceleration. In this respect, it’s purpose is exactly the same as the more traditional over the shoulder wrap. The different places the wrap ends up are just the result of differences in the shape of the swing path.

When players develop more hand and arm rotation, they should be ending up more across the body, but it should happen naturally if the swing pattern is correct and the hand and arm are relaxed. Emphasizing the left wrap and trying to make it happen will usually have negative consequences. Typically when players try to mechanically force themselves into the wrap position, they end up reducing their swing extension, hitting through the ball less, and losing racket head speed. It’s frequently a disaster for competitive junior players.

With Federer, you see a range of hand and arm rotation. Sometimes he rotates only slightly further than a classical forehand. On these balls he also makes the universal finish position. Itt looks like the extreme players at the same point in the swing.

When the hand rotation is extreme, he doesn’t make it all the way there, again, the same as the extreme grip players. In this case, when the hand gets across to the left edge of the body, we can see that the spacing between the racket and the torso is definitely less and that the hand is noticeably closer in to the body. The angles in the arm position are also different. Sometimes the upper arm doesn’t reach parallel with the court and is still tilted slightly downward. The forearm is also at a lower angle, at most about 30 degrees to the court surface, but sometimes less and even at times horizontal.

These are the correct technical checkpoints, but again, it’s important to recognize that extreme hand rotation is a variation on a continuum and not the absolute norm. Again, people who try to blindly model Federer without examining the full range of his technical options end up exaggerating the left side finish and lose the ability to really drive through the ball.

Sometime when Federer rotates the hand and racket all the way, the extension along the line of the hit is considerably less, but that’s not always the case. It depends on the hitting arm position he uses. Usually, Federer uses the double bend position when his rotation is extreme and comes off the ball more quickly. But when he pairs the extreme rotation with the straight arm position, the extension, can be as pronounced as on any forehand he hits. Again, it’s that complexity thing. Too many players copy the extreme, and end up overexaggerating it to the detriment of their strokes.

Does Federer gain an advantage by pairing the conservative grip with extreme rotation compared to the players who are more underneath the handle? An absolutely definitive answer will require quantifying the path of the racket and the speed of the swings in pro tennis. Then we can compare classical and extreme swings and measure the differences in the shapes of the swings and in the speed of racket head from player to player and shot to shot. Currently we doing some new analysis of spin rates in the games of several top players, including Federer, which will shed at least some light on these questions. I’ll be writing more about this in the Heavy Ball section of the site.

But from the study of the high speed video, I believe this combination is what allows Federer to hit with great pace but at the same time generate surprising amounts of spin in relation to the other top players. More hand and arm rotation gives him the ability to hit wide spin variations, but he is still able to hit through the ball more directly than the extreme players because of his grip.

Again it’s the best of both world’s. He can generate more pace with less effort due to the mechanical structure of the eastern swings, but by mixing in the hand and arm rotation, he can combine raw ball speed with spin rates that equal or possibly even exceed the extreme grip players. An eastern grip that produces heavier ball rotation than the extreme players. If you watch his matches, you have to agree that he is producing effortless power, but also a far greater variety of balls on the forehand side than most of the other players.

One way the classic grip may actually have a technical advantage in producing spin is that it allows the player to actually rotate his forearm further backwards at the start of the foreswing. We saw that with the extreme grips, the arm was naturally rotated back, with the forearm turned upward toward the sky. Agassi and Sampras had less forearm rotation backwards with their more conservative grips. But this assumes that the shaft of the racket is pointing basically straight backwards and is more or less parallel to the court.

What’s different here in Federer’s motion? Amazingly, with his grip, he has the option to rotate the forearm further backwards. When he does this, the tip of the racket drops at an angle significantly below the ball. In fact the racket tip can point down at a 60 degree angle or so to the court, and appear to be two or three feet below the height of the ball. You see this to a certain extent in all the players, especially on really low balls and really short balls. But the high speed footage shows that Federer does it on some higher balls, up to around waist level. It’s not on all balls.

On his basic drives, we see the forearm in a position similar to a player like Sampras, with the racket shaft basically parallel to the court. However, on some balls he rotates the forearm back much further. What does this mean when he swings the racket forward and rotates the hand and arm? The racket moves further upward on it’s way to the contact. The result is probably more racket head speed and more spin, all generated automatically in the course of the swing. It’s important to note that this is all done without any change in the hitting arm position itself, that is, the internal relationship between the racket, wrist, and forearm. They simply rotate backward as a unit, and then forward, again as a unit, over the course of the swing. The length of the rotation of the “windshield wiper” is naturally lengthened.

The Myth of the Wrist

And what about the role of the wrist in all of this? Federer’s emergence has stimulated another round of impassioned, if repetitive, debate about the movement of the wrist on the modern forehand, and it’s role in generating speed and/or spin.When you look closely at the high speed footage, though, you’ll conclude that it’s no different for Roger than any other player. When he is in the double bend position, the wrist is laid back before, during, and after contact.

As we saw, with the straight arm position, the wrist can actually more laid back. There are also balls hit with extreme rotation and/or hit on sharper angles where he goes in and out of the double bend much more quickly—again like the other players we’ve looked at. Some version of this laid back position is still the norm even with all the advances Roger Federer may have achieved in technique on the modern forehand. Let’s not confuse the rotation of the hand and arm, and what happens to the wrist during the followthrough, with the position of the wrist at contact.

Racket Tilt

Another controversial question raised about Federer, and the other top players for that matter, is whether they “tilt” the top edge of racket forward at contact (For one view on this, read Kerry Mitchell’s article which deals with racket tilt from the point of view of the different grips. Click Here.) Does “racket tilt” increase topspin? Does it have an effect on ball speed and/or the trajectory of the shot? Is it even physically possible? One of the very few points that most teaching pros probably have agreed upon over the years is that the racket has to be perpendicular at contact to produce topspin. I’ve always believed that was critical in teaching topspin on the groundstrokes—and I still do. Is there now another option?

There is no arguing with the fact that forward racket tilt does exist, at least at high levels in the pro game. I first saw it in the high speed filming we did at the U.S.Open in 1997. I was showing Dennis Van der Meer a frame by frame sequence of Pete Sampras hitting an incredible, ultra high topspin backhand. The contact point was actually above his shoulder level, yet Sampras still managed to hit up and over the ball. Dennis was the one who noticed the angle of the racket face. Then, with a wry smile he said something like: “John, I think what we are seeing here is impossible.” When I looked closely I couldn’t believe what I was seeing either. Pete’s racket face was closed or “tilted” about 30 degrees forward.

Over the years in the high speed footage I’ve found multiple other examples of the same thing. I’ll have to say at this point they appear to be a variation, not the rule. Interestingly, probably most of them have been on high topspin backhands. But there is no doubt it happens on forehands too and you can definitely see it in the Federer footage. The racket face tilts a few degrees forward on maybe around 20% of his forehands. And the ball certainly doesn’t go down into the net, as most teaching pros (myself included) would have predicted. But exactly how does it affect the shot? If it does have some positive effect, could it possibly apply to lower level players, outside the rarified air of the stadium courts of the world tour?

At this point we don’t understand the physics of it. But I’m excited to report that is about to change! This is because the noted Australian physicist Rod Cross has agreed to take on the topic for a future Tennisplayer article. Does tilting the racket head depend on some combination of ball speed, ball height, and swing trajectory? Stay tuned and hopefully we can start to develop a factual basis for intelligent discussion.

What Can We Learn From Roger?

Which brings me to one of the questions I posed at the very start of this analysis. What can we learn from Federer? Are there things for every player to emulate? In this section, we’ve included a few more click through movies from the high speed video to help you ponder those questions.

I already know what is going to happen to a lot of players after reading this article. They will immediately head to the court and start trying to copy Federer’s extreme arm rotation. And a lot of them will go for the racket tilt as well, without even waiting to read Rod’s article. What ever is most idiosyncratic or unusual—that’s what many players want to copy—and maybe in their heart of hearts they don’t care that deeply about whether they are really doing something that could improve their competitive results.

And that’s OK! It’s still a (mostly) free country, and tennis is supposed to be fun. If you want to dream you are Roger Federer, you are certainly allowed. Just don’t expect that your dream will give you his forehand. Because chances are quite high that it won’t. And don’t blame me if it doesn’t. What I am trying to do is understand and describe his technique, not advocate that everything he does should be a universal model. And that’s an important point—at least to me—and I find a lot of players miss it when they read these articles.

Don’t take this the wrong way. I do hope these articles inspires you—they did me! And what you get out of it is obviously yours to take! But I also have some ideas about what is reasonable to try and apply. The elements in Federer’s forehand that would be most beneficial for most players to copy are, ironically, the ones he shares with the other players. These are in the preparation phase, specifically the unit turn, the left arm stretch, and some version of a compact backswing.

Time and time again I get emails or phone calls from players, or they come to our tennis school and they want to improve their forehand by developing advanced techniques. And then we film. And in 90% plus of the cases, they don’t have the basic foundation to incorporate these more advanced elements that they are convinced are the key to their improvement. That thought may not sound too glamorous and exciting, but it is the god’s truth. I’ve seen it in 3.0 players and I’ve seen it players who are in the top 100 in the world—and I’ve seen it everywhere in between.

So that’s number one—please don’t “get your racket back early.” Learn to start your preparation with your feet and body. After the preparation, the next critical thing is the basic, double bend hitting arm position. Then the universal finish position, regardless of your grip. And you can find all of those in Federer’s forehand. If his forehand teaches us one thing, it’s the need for meticulous discipline in perfecting the basics. But what about all the wonderful new stuff? Why bother even trying to figure out the marvels and complexities of his forehand in the first place if you are never going to try them?

One answer is that the satisfaction of the understanding alone is worth the ride. In a weird way I feel I’ve had the privilege of playing my own little 5-set match against Roger—just him and me battling it out ball after ball in the struggle to understand the most advanced weapon in the modern game. For the last few weeks when people asked me how I was doing, I’d reply; ”Roger Federer is kicking my ass.”

It may sound strange, but I developed this eerie sense of just how tough it would really be to face him across the net. Just when you think you’re getting a handle on what he does, he throws you a whole new range of impossible combinations. He may seem reserved, but believe me he is a rough customer. If you have the nerve to go up against him, get ready for some tangible pain. I won’t go so far as to say that I came out the victor in my little imaginary match, but I guess I feel that I did hold my own by sticking with him long figure out something some of the incredible things he can do with that forehand.

But I didn’t do it just for the pure understanding. I sincerely believe that every player should experiment to see what they can learn from an innovative player who will also likely end up with one of the greatest records in tennis history. Depending on where you are honestly are with your forehand technique, that experiment could include different hitting arm positions, and variations in torso and hand and arm rotation. Find out what you can make work in your game! Just don’t send me any video of you with an extreme left side finish but no shoulder turn in the preparation!

That’s one of the main reasons I write these articles—to find these things out for myself. When it comes to the Federer forehand, I found an incredible benefit in understanding how the hand and arm rotate and how this can be done with an essentially eastern grip (the same grip I hold). I’ve incorporated that into numerous situations—hitting high loopers with more topspin, and also short balls, crosscourt angles, and even running down the line passes, some of the things I have struggled with most over the years. Learning to rotate my hand arm and racket as a unit has made a tangible difference in what I feel I can do with the ball. I’m inspired and excited! And I can visualize hitting a few more passes at critical times against my old friend and practice partner, Mr. Scott Murphy. We’ll see.

I also benefited from my experiment with the head position. I couldn’t come close to turning like Federer, but the process actually helped me focus on the key point of keeping my head still at contact. And I experimented with the straight arm hitting position combined with more torso rotation, especially on high inside balls, since I prefer to play inside when I’m on the baseline. But I’m working to incorporate the head position and the hand rotation first before I decide about trying to actually incorporate that.

I’m not saying you should follow these exact steps (although they are good ones!), but I think you get the point. That’s the beauty of having this incredible new high speed resource. We have the potential to really see what is happening instead of just speculating. And that reality forms the basis for making productive changes--basic changes as well as more advanced ones. Not to mention just the pure pleasure of understanding Roger better when you watch his matches—and maybe having a little bit of sense yourself of what life is like on the other side of the net for his opponents.

John Yandell is widely acknowledged as one of the leading videographers and students of the modern game of professional tennis. His high speed filming for Advanced Tennis and Tennisplayer have provided new visual resources that have changed the way the game is studied and understood by both players and coaches. He has done personal video analysis for hundreds of high level competitive players, including Justine Henin-Hardenne, Taylor Dent and John McEnroe, among others. In addition to his role as Editor of Tennisplayer he is the author of the critically acclaimed book Visual Tennis. The John Yandell Tennis School is located in San Francisco, California.

Tennisplayer Forum Let's Talk About this Article! Share Your Thoughts with our Subscribers and Authors! Click Here

Contact Tennisplayer directly: jyandell@tennisplayer.net

Copyright Tennisplayer 2005. All Rights Reserved.