Your Strokes:
Kyle Doppelt Serve

Analyzed by John Yandell

Some of our subscribers know that I am a proud graduate of Yale University, but my Ivy League loyalties are now divided after my work with Cornell varsity player Kyle Doppelt. Kyle (and his charming mother Diane) spent a week in San Francisco earlier this summer. Besides taking them to Shalimar, the world's greatest Indian/Pakistani restaurant, we made a lot of progress on Kyle's attacking game.

Kyle moved to a more conservative forehand grip--the Federer 3 1/2 / 3. He found this allowed him to come forward on his returns and also to play up on the baseline off the ground. This meant he could better take advantage of his natural attacking instincts and solid technical volleys.

But probably the biggest changes we made were to his lefty serve, which he told me at the outset was his best shot. Kyle is tall and he hits the ball a ton. The first thing I looked for in the video was the pro racket drop position and Kyle had that naturally--the sign that he had the kind of shoulder flexibility it takes to develop a very high level--maybe even world class--motion. He also had a very effective wide serve in the ad court. It may have been his most effective shot, but his serve was the motion we ended up changing the most.

We were really lucky during part of the time Kyle was here that my good friend Jeff Salzenstein, Stanford All American and top 100 tour player, was also in town. In addition to working on his backhand and backhand return, (see it later in Your Strokes!) Jeff was kind enough to hit with Kyle a couple of times. The first day at the beautiful Harbor Point Club in Mill Valley, Jeff was virtually mobbed by a ladies doubles group all of whom remembered him from the glory days at Stanford. Apparently, they wanted to introduce him to a few of their daughters, but that's a different story. (Jeff, by the way, claimed this happened to him all the time.)

More relevant to our focus here, is that Jeff has a high level world class lefty serve. I have video I filmed of him at the U.S. Open, hitting 130mph on the radar gun on the grandstand court. So it was natural that Kyle paid close attention to what Jeff was doing and immediately adopted him as a model. We also studied the Sampras high speed footage, which had in turn served as a model for Jeff over the last 2 years. It was a great synergy for all three of us.

So what was the rationale for changing a player's best shot? Well, "best" doesn't necessarily mean maximized. After Kyle served about 3 points, Jeff told me, "It's way too flat." And he should know. I'd had to laugh because I was thinking the same thing. There are two problems with hard and flat. First at higher levels players will get onto the speed and turn it into an advantage on the return. Second, under pressure it's really tough to serve high percentages. Interestingly, it was the same process Jeff had gone through himself in the work we had done over the last couple of years--changing his own best shot in search of a heavier and more consistent ball.

When we looked closely at Kyle's motion on video we could see what Jeff was talking about. Kyle's toss was quite far to his left, and so was his contact point. As you can see in the animation, his racket was virtually straight up and down at contact. That made it virtually impossible to add the topspin component we associate with the "heavy" ball. (Click Here.)

Related to this was his stance, which was an extreme pinpoint. Kyle started with a super wide stance. This made it necessary to slide his foot to his left as the motion unfolded. But in Kyle's case, this sliding movement with his left or rear foot was quite extreme. It ended up continuing to the left past the position of the front foot.

We've looked at the complex issues regarding stance in detail before in the Advanced Tennis series. (Click Here.) As is typical with the extreme pinpoint pattern, Kyle's stance was causing him to open his shoulders too soon at contact, and restricting his knee bend as well. Once we saw this, we turned to the video both of Jeff and of Pete Sampras to create a new model. One of the fun parts of having Jeff there, was that he could demonstrate the result right there at Harbor Point, in real time, at 125mph. That was a pretty strong selling point.

As we seen, differences in the amount and type of spin relate to relatively small differences in the toss position and in the path of the racket to the ball. Basically you can see the amount of topspin in the angle of the racket shaft around contact. If the racket is basically straight up and down then the only real spin the player can generate is slice. The more the racket is angled, with the head tilted slightly down toward the server's head, the more topspin. But the toss, the racket position, and the stance are all interrelated. The further the toss is to the side, the more straight up and down the racket, the more open the shoulders at contact, and the greater the need to slide that foot to get in position to hit the ball.

When Kyle and I examined all this on video, his response was fabulous--he absorbed the new elements almost instantaneously. We narrowed his stance until he was comfortable keeping the back foot in position, and at the same time played around with the ball toss until he found a contact point further to his right that was comfortable for his motion. He never looked back, and his serve immediately looked better and heavier. He took to the whole process the way our digital still photographer J. Gregory Swendsen takes to a rare rib eye steak in Indian Wells.

Note for all three players the ball position is inside the hand.

Which brings us to an important point about using visual models. The images of other players can demonstrate the important elements. But we shouldn't think of the models as chiseled in stone and completely rigid. The model functions as a blueprint or guide, pointing the player in the right direction so he can feel how to make the elements work for himself.

Jeff initially modeled Pete's motion, but ended up owning it for himself, with some subtle differences in the spacing of his starting stance, the alignment of the feet and shoulders, as well as the total amount of body rotation. And it was the same for Kyle. If you look closely at their starting stances you can pick out the differences.

But there are some core elements all three share in common that allow them to add the topspin component to the serve, and also, in my opinion, use their legs and torso rotation patterns more efficiently in the motion. You can see this in the still photos at contact. First, the position of the ball toss is inside the edge of the hitting hand. This is the key factor in generating topspin. You can see how this happens in the angle of the racket. The tilt or angle shows that the racket is still rising at contact. For Kyle this is a clear difference, compared to the animation of his serve before the changes.

The second factor is the angle of the torso, which is closed, or at an angle to the baseline. If the torso rotation is too fast, so that the shoulders are too open at contact, the energy from the legs and the upper body doesn't, in my opinion, pass as completely or efficiently upward into the hit.

So far Kyle's results have validated this theory and the work he's done. He spent a few weeks training hard in Florida at Bollettieri's, and also won several matches in summer tournaments there for top college players. I can't wait to see how his season goes for the Big Red up there in Ithaca, New York. I won't be writing this into the alumni magazine, but when they play Yale, I'm probably going to have to root for him, even if the team match is on the line.