This page is a public service we do at S3. It's about racket sports and how to improve at them. But in a sense it also has something to do with data science and simulation science as well. S3's founders are active in racket sports communities at the local, regional, and national level. We believe these sports are a good way to keep the body and mind fit; a prerequisite for doing good science by the way. And our perspective on improving at racket sports is based on modeling. Some would say it's about developing a "system" that works, but there's scarce little difference between a model and a system in that sense.

We've worked with some of the most respected racket sports players and teachers in the world. In addition, we sponsor tournaments, host youth programs, and do a great deal of work to train up-and-coming players in various sports. We think that--coming from technical backgrounds--we have a fairly unique perspective on how to improve, and can make a significant contribution to the way these sports are taught. So this part of our web site is a resource for people who want to play better racket sports. Before reading on, the reader should understand that the most important thing in becoming a better player is to love the game. One of the most tragic things we see is kids being forced to learn a sport because their parents think it's good for them even though the kid can't stand it. Sometimes it's the "tennis parent" who tries to drive the child into tennis from an early age so they can later someday earn a college scholarship. Less frequently, we've seen the same thing in squash. We've seen kids tortured by table tennis lessons to the point where they are regularly in tears or even vomiting. Adults are not immune either. We've seen boyfriends and girlfriends dive into a sport solely for their partners, and it often tears the relationship apart. You've got to love it, or it will always suck to practice and it will always be painful to absorb the stress of play. But if you do love your sport, you not only can take the stress, you can feast on it. To help prepare the table for that feast, we provide some tools to: 

  • Model player performance: How to evaluate the level of an opponent, or, more importantly, how to self-evaluate your own level. 
  • Model player improvement: How, starting with your assessment of yourself as a player, you can get better.

To be sure, you will likely want relative and absolute models of player performance. For instance, if you are comparing your performance with an opponent, you'll want to measure both on an absolute scale: "0n a scale of 1-10, my speed is a 7 and my opponent's speed is an 8, so I probably won't benefit from challenging him on speed". But when it comes to improvement, you'll want to use a relative scale: "I may be slower than I used to be, but I'm 67, and have medical issues with my knees. I'm as fast as I can be under the circumstances (10 out of 10 on a relative scale) so I will want to focus my training on other aspects where I can still improve. Work on the weak spots." 

When S3 founder Jamie Lawson was in his late teens, one of his racquetball partners was Jimmy Doyle, a world's top 20 player at the time (they still play together, more than 40 years later). Racquetball was booming back then, and there were lots of books on the sport coming out at the time; too many to read all of them. Jamie asked Doyle what he thought the best racquetball books were. Doyle told him: "In my opinion, the best racquetball book is Timothy Gallwey's The Inner Game of Tennis". Puzzled, Jamie asked "how can a tennis book help racquetball players?" "Outside of some mechanical differences," Doyle answered, "most racket sports are pretty similar. Certain athletes may excel at one sport or another, but the concepts that must be mastered are much the same." Our discussion here is an attempt to enumerate and understand those common concepts, and we think that by making the discussion independent of any particular racket sport, it's easier to focus on the bigger picture, and improve at how to play your sport. We think this is a missing perspective. Something we will do from time to time in this discussion is insert little boxes in the text with general truisms about the racket sports.

Racket Sports Truism

You get better faster if you work on your weaknesses, rather than your strengths.

A general theme in our approach to improvement--as you may have already guessed--is to work on the weak spots, and a supporting theme is to use the strong spots to improve the weak spots. This seems like an obvious approach, but often, players follow Molly Ivans' rule of "dance with them what brung ya", They think, "I worked on my cardio and it got me to this level, so I'll do more cardio and hopefully it will get me to the next level". But champions are always working on different stuff. A reality of two-player or multi-player games is that the opponent will seldom play into your strength, and will usually look for your weak spot and try to exploit it. So whatever the short poles in the tent are, figure out how to make them longer. But just as important is to realize that there are more poles in the tent than those you can see. This is a good reason to work with more than one teacher. Each of them will show you different tent poles. But there are always pieces of the game you're missing. And you may lose because someone else recognizes those parts of the game. Our next section, on the model player, is about discovering new tent poles in your game. Working the referee is a good example. Lots of players see the referee as a bystander who won't be manipulated by the players. That's not how real matches work. In his autobiography Open, Andre Agassi describes one of his first tournament finals, as a junior, where the rule was that if you didn't have a referee, you call balls on your side of the court in or out, and those calls cannot be challenged. At match point (in a strange scoring system), the winner of the point would win the whole tournament, Andre hit a winner that was in the court by several feet and therefore unquestionably in. But his opponent, understanding this part of the game better than Andre, simply called the ball out and collected the championship to the boos of the fans. That doesn't mean you should call balls out just to benefit your score, but instead maybe demand a line judge at the start of the match to remove that tool from the opponent's arsenal, and learn to recognize when the opponent is trying to work the referee and also learning how to work the referee yourself to keep the referee centered.

Blair: The Model Player

Let's begin with a typical pattern that players go through in the course of their improvement. We'll call our player Blair. Blair could be male or female, it doesn't matter. One day, a friend invites Blair onto the the court and goes over the rules. At this point, the rules are the game. They play a few games. Blair has fun, but doesn't win anything, and feels kind of lost on the court. It's like a wilderness adventure hike and Blair has no idea where it's going. After the first three or four outings, Blair finally strikes a victory, and feels more at home on the court, without as much need to look down at the lines to get a location fix. At this point, the court is the game. Blair starts playing other hackers. Blair beats some of them consistently and loses to some of them consistently. Blair starts practicing more, and in order to turn some of those losses into wins, maybe buys a better racket and better shoes. At this point the equipment is the game. There's only one tent pole at this point and it is the lines painted on the floor. Suddenly, Blair's on the contacts lists of most of the hackers at the club, and some improvement results from regular play. But not long into this, Blair reaches the first plateau, and improvement slows to a crawl. 

Blair doesn't really know how to get off the plateau. Lessons might help, but what is a teacher going to do except say "get out and practice"? Not only that, taking lessons is like giving in and admitting that you can't do it on your own. But Blair has noticed that the better players are typically in pretty good shape. And they have to be to get away with wearing some of those spandex workout outfits. Blair has walked passed the gym a hundred times now between the locker room and the courts. But one day Blair makes that left turn and goes into the gym. With a modicum of commitment and a couple months work, Blair's bench press and squat maxes are up about 40%. Blair steps lighter on the scale now, and Blair's checking account is also about a thousand dollars lighter on account of all the new spandex workout clothes. About this time, Blair's game is also seeing a bit of a bump. Blair feels a little less depleted after an hour of play, and has started to take home some additional victories. At this point, Blair sees the game as an athletic event. So athleticism is another tent pole. And Blair's name is now on the contact lists of some of the players on the club's intermediate ladder. But Blair consistently loses to them, and it's starting to feel like another plateau. 

It's time! Blair makes a call to the club's teaching pro, whom Blair refers to respectfully as "Coach", and to which the teaching pro lightly blushes. They hit the ball around for a bit and then the teaching pro raises a racket and says "okay, I see what the problem is here. It's in your technique, your mechanics." The teaching pro--seeming to be a font of knowledge--sets Blair's feet and guides Blair's arms through the swing. "The correct swing," says our muse, like a character from Alice in Wonderland, "has definitive starting points and ending points, and definitive points in between. start 'here', then go to 'here', 'here', 'here', 'here', and finish 'here'". At this point, the game is the swing. And it is as complex as a fine wine. Blair practices the mechanics relentlessly, and is sometimes queried by the hackers on such stuff as "which grip to use for which shot?" But Blair is still getting beat in Podunk hit-around tournaments, and by players who aren't even changing out of their warmups, which is good because they'd be a sight to be seen if they tried to fit into spandex workout clothes. That's disappointing!

It's another plateau, but all is not lost. After an emotionally crushing defeat to a player with an inferior swing, Blair collapses just off-court, and an old codger hobbles over, taps a crooked stick on Blair's shoulder and says, "you know, with that forehand, I wouldn't lose a rally." Blair recognizes the old codger from pictures on the clubhouse wall; pictures of a better day when justice ruled and giants walked the earth. "I thought your swing would be...you know...perfect", Blair replies. And then the old codger looks left, looks right, then looks Blair right in the eye and says: "That's the thing, it's not about the swing!" Puzzled, Blair says "but but but I'm confused". And the old codger says: "Well, if you want to understand, see me here at 7AM tomorrow."

So 7AM comes and Blair arrives to a frightening sight: the Pantheon. This is Blair's "Kung Fu Panda moment", you know, the one where Po realizes that he will be training under the direction of Master Shifu, with the Furious Five: Mantis, Monkey, Viper, Tigris, and Crane. Blair has seen them before, but only at tournaments where viewing was hard because it was standing room only. After a great deal of clubhouse talk and jabber, Blair finally get to play Master Shifu. At the end, Master Shifu closes with a question: "All of your shots were better than any of mine, and yet I won all of the rallies. How do you explain that?" This is the moment Blair realizes that the teaching pro taught by telling, and Master Shifu teaches by asking questions. This is also the moment that Blair finally realizes the difference between a teaching pro and a Coach. 

Blair spends many days and many nights searching for an answer to Master Shifu's question. He asks each of the Furious Five and they either don't know the answer or won't say. Finally, in a dark dream it comes to Blair, and the next morning Blair greets Master Shifu at the club and says: "Coach Shifu. I have an answer to your question." Master Shifu waits anxiously for the answer and Blair continues: "It's not about the swing!" Master Shifu smiles approvingly. At this point, the game is about tactics: how to move the opponent to create opportunities; how to confuse the opponent with deceptions; how to use the entire volume of the court as a creative space. At this point, Blair also realizes that the most important place at the club is not the court and it's not the gym, it's the clubhouse where Blair and the Furious Five hold discussions on these important matters. Without the clubhouse, the club is just a place to play.

Blair occasionally takes a game from some of the Furious Five, but every time he plays one of them Blair's palms sweat. Blair tightens up and misses a lot, feeling exhausted even before they play. Crane lets the secret out: "It's how it is in life," says Crane. "So long as I'm more comfortable with the stress than you are, I will win." This is an epiphany for Blair. At this point, the game is the competition. It's how each of the players deals with stress. And Blair works on personal realization, on brushing off stress and just playing the game. Finally, one day, at the end of the match Master Shifu stands defeated. "How did you do that?" Master Shifu asks. "I don't know," says Blair. "I had a bunch of difficult situations, and in each of them, I saw the shot I wanted to hit and new I could make it." Master Shifu smiles, and says "now you know how to play!" At this point, Blair returns to the locker room, and trades in the spandex workout wear for a crooked stick.

One takeaway from this is that while a player may reach multiple plateaus in their journey, the plateaus are different. They come about for different reasons and they require different tools to overcome. Eventually the player reaches age-limiting plateaus, where they have technique, game, competitiveness and even confidence, but basic biology imposes a reduction in their athleticism. These players can be the most instructional for young players to watch because their games are distilled down to just the parts the younger players don't already have and need to build. At the time of this writing, John McEnroe presents a great example. McEnroe's athleticism has faded, and the varnish is even peeling on his technique--no doubt giving in to the problems of older joints. But at the tactical level, he still has an effective approach and his angles allow him to prevail against highly skilled players a half or a third his age. 


Let's begin with the "start, here, here, here, here, finish" example, that the the teaching pro went through with Blair. Blair made a mistake in assuming that this was the "perfect" swing. But Blair later learned to think of it as an example of a swing, one Blair could generalize from and could train from to have a better swing. One of the things we do in science is to generalize from examples. The name for that is "abstraction". That is, we develop general patterns or models that explain many examples. As we've already made clear, models are important to us as scientists, and models are also important to players so that they can build cohesive play. And we are always looking for models at higher and higher levels of abstraction, because higher levels of abstraction are more general, meaning that the models explain more examples. 

It turns out that abstraction is important in racket sports as well. So if you let the details of "start, here, here, here, here, finish" dominate your conceptualization of proper swing mechanics, it will do more harm than good. Recognize that in match play, you will seldom be able to achieve those positions. All racket sports are, as tennis pro and author Peter Burwash says, "games of emergency". In fact, another way to think about Blair's teaching pro's example swing is that this set of positions is what you want to prevent your opponent from achieving. That's what creates the emergency for your opponent. And your opponent will certainly do his or her best to create that same emergency for you. Drill with those positions in mind, but not locked in conceptually. Once you are able to produce a swing, in practice, that flows through all of those positions (more or less), try to find the right abstractions that explain what those positions were trying to achieve. Instead of "your hand should be here and your racket should be here", it may be "the face of the racket should strike the ball flat, and the inertia should drive the ball toward the target". That's one level of abstraction. After we master it, we can look toward higher levels of abstraction, like, "the proper swing produces a ball that moves with sufficient pace and within a certain margin of error of the desired flight path". This abstraction applies to any ball, not just the one that you have ample opportunity to setup for and which allows you to flow through each of those six prescribed positions. So maybe you're on the run and the ball is fast moving away from you, and "start, here, here, here, here, finish" is simply not possible. But it is likely still possible to think about how to balance your swing to hit with sufficient pace, within a certain margin of locational error. At a still higher level of abstraction, you might conceptualize that "the proper swing produces a ball that creates an emergency for your opponent". Being able to hit with sufficient pace and accuracy may be how you create that emergency, but the real goal is not to hit with pace and accuracy, but instead to create the emergency for your opponent. 

Players often talk about "going up a level" in their games. On the surface, they think of the levels as numbers, like going from a 3.5 player to a 4.0 player. And that's a reasonable way to measure competence, but more often than not, what the player is actually doing as they are moving up is reaching a higher level of abstraction in their game. So the real goal as you seek ever-continuing improvement is not to repeat the "start, here, here, here, here, finish" routine, but to master ever higher levels of abstraction of what "start, here, here, here, here, finish" really means. To be sure, you should not always try to think at the highest level of abstraction. If your mind is focused on "maximize the opponent's level of emergency", but you don't know how to do "start, here, here, here, here, finish", you are likely to play foolishly. A good rule is that when you play, focus on the highest level of abstraction you have mastered, and when you practice, focus on the next level up from there. So long as you practice regularly, this system will avoid most plateaus. 

So there are a number of takeaways in the diagram above. 

  • Each level is one way to accomplish the thing in the level above it. So "start, here, here, here, here, finish" is just one way to strike flat with inertia toward the target. There are other ways and you may have to figure out another way when, for instance, you're on the run and in extremis. 
  • Each level generalizes across sports better than the level below it. So the body positions in "start, here, here, here, here, finish" might work for racquetball, but it might be an inferior swing for paddleball even though the two sports are kissin' cousins. With racquetball's faster ball there's less time so the swing tends to be shorter. The ball is slower in paddleball, so a bigger swing may not only be possible, but also necessary. Both sports favor a flat swing so "strike flat with inertia toward the target" is still a good pattern. But it doesn't extend to squash where underspin is desired on both the forehand and the backhand, in order to bring the ball quickly to the back court and then die before reaching the back wall. However, "sufficient pace within a margin of locational error" is an apt description of swing mechanics for racquetball, paddleball, and squash, as well as others, and you'll do well if you can master "maximize opponent's level of emergency" in pretty much any racket sport.
  • Since our approach to racket sports attempts to favor no specific sport, our discussion will do its best to remain at the higher levels of abstraction. Say, the green and above in the chart, occasionally dipping into the yellow and red levels for examples.

Predictive Versus Causal Models

In the sciences, we tend to have two different kinds of models:

  1. Predictive models tell us what to expect,
  2. Causal models tell us why to expect it.

Say we are trying to calculate the depth of water wells by dropping stones into the wells and measuring the time before we hear the "plunk" of the stone hitting the water. We could start with a dozen wells of known depth and drop some stones into them, then develop a regression model with a couple variables like time-to-plunk, time-squared, ambient temperature, sign of the zodiac, etc.. And it might be a pretty good model, but it would do nothing to explain why we should believe the model. It's a purely predictive model. Google developed a system for predicting where the flu would break out next based purely on Google search queries and the locations they were made from. The system was called Google Flu Trends. And it turned out that Google Flu Trends was able to predict flu trends as well as the Center for Disease Control could with their clinical data. The Google Flu Trends model had lots of crazy stuff in it, like whether you were querying about blue cars or red cars which in all likelihood actually had nothing to do with flu trends and there is certainly nothing to make us believe in the model. "But we don't need to know why. It just works, hurray! Trust us."

In contrast, we could start with a model of how a stone falls in a field of constant gravitational attraction and build a model that way. That is a causal model. It comes with an explanation of why we should believe it; why it works. The purely predictive model may actually give better predictions. After all, that's its job. And where the (simple) causal model doesn't take into account things like buoyancy and air resistance, the predictive model implicitly incorporates those things because of the way we built it; because those things were present in the data we used to build the model. But therein's the rub. We have few clues about how to improve the purely predictive model, whereas we can improve the causal model by analyzing it and seeing what causal contributors it incorporates and determining which causal contributors it doesn't incorporate and how to incorporate them. In fact, that's what happened with Google's Flu Trends model. After they had published a whole bunch of papers on how good it was, the model started to "drift" because a lot of the stuff that made its way into the model was junk. And there was much embarrassment amongst the people who had written those papers, and the CDC went on predicting flu trends with reasonable accuracy using a causal model that they had scientific reasons to believe in. 

So now let's get back to racket sports and see where this applies. A player may think that wearing their lucky turtle necklace increases their performance on-court. And they may have collected some data to support it: they've played five matches recently, three with the turtle necklace which they won, and two without, which they lost. What could be more conclusive? But correlation is not causation. Perhaps the three matches they won were against short right-handers and the two matches they lost were against tall lefties. And this might speak much more to the cause of the victories and losses than the turtle necklace. Statisticians call the turtle necklace theory "spurious correlation". This is why causal models are more difficult to build, because you have to rule out spurious correlation. The important thing is that the lefty-righty theory provides a viable causal explanation of performance whereas the turtle necklace does not. The lefty-righty theory gives us reasons why we should expect that the model is a good predictor of outcomes. It might not be valid. It could be that the two lefties were 5.0 players and the three righties were 3.5 and under players. That might have more to do with it. And when we start getting beat by right-handed 4.5 players (providing evidence that it's not a lefty-righty thing), we should adjust our causal model accordingly. 

Going back to "start, here, here, here, here, finish", on its face, that's a predictive model. Your coach is saying, in essence, "swing like this and you will win, trust me." As we move up through the different abstractions of the game to higher and higher levels of play, our models of the game become more causal. Each level of abstraction offers some explanations of why we want to start "here", and why we want to pass through "here, here, here, and here", and why we want to finish "here". While the model becomes more abstract, the understanding becomes more concrete. Consider a topspin tennis swing (where the red and yellow abstraction levels will be different than those stated in the figure, because the figure assumes a flat swing). The predictive model says "swing up to the ball, contact it high on the strings, brush it with the strings, release low on the strings". That's kind of the "here, here, here, here" of a topspin tennis swing. Meanwhile, the causal model says "rolling the ball on the strings causes it to spin. The nap (fuzz) on the ball catches the air and gives the ball resistance to the air. As the ball moves toward its target, it essentially creates a wind around it. In the case of topspin, the top of the ball is spinning into the wind and the bottom is spinning away from the wind. So the force of resistance to the wind is greater on top than on the bottom. The greater force on top pushes the ball down faster than gravity alone would take it, allowing the striker to hit with greater pace and still keep the ball in the court." It's hard to move from the predictive model to the causal model because there's a lot more there to understand. A big part of growth in your sport is making that move. Ultimately, coaches and teaching pros may teach the predictive models to beginners for simplicity. Predictive models are typically easier to understand. But the coaches need to (and usually do) understand the causal models of what they teach. When you start asking "why?" in your mind, your brain is trying to move from a predictive model of the game to a causal one. That's part of the reason real coaches try to make you think about the "why?" questions. In the rainbow chart above, you can think of each level of abstraction as being more causal. It answers the "why?" question raised by the level below, and when you can answer those questions, your game goes up to the next level.

A Model for Measuring Players

Our approach to racket sports--our abstraction--is a system for measuring players. It's a system that players can use to measure themselves, and also to measure potential opponents. In this section, we describe the system for measuring players. In later sections we will build on the system so that we can use it to improve players and exploit opponents. Our system is not entirely new. It's rooted in something Bill Tilden said in How to Play Better Tennis back in 1950:

Tennis matches are won or lost by the sum total of physical condition, courage, intelligence, experience, and stroke equipment of a player. If your sum total is greater than that of your opponent, you win; if it's less, you lose. 

This idea came to us in a more usable form through former tennis pro and now tennis coach Dan Emmerson, and with some nuance from racquetball hall of famer Charlie Brumfield who says "It's not about ball striking, it's about everything else". Dan Emmerson's riff on this idea is to divide the player into five equal parts, measure each part on a scale of 1-20, and sum the total. While Dan attributes this to Tilden, we can't find anything more specific than what Tilden said in How to Play Better Tennis. So we credit Dan for this abstraction. And based on 70 years of progress since Tilden's book (and a lot of brow beating by Charlie Brumfield), we break down the five pieces differently than Tilden did:
  1. Athleticism: The player's fitness, speed, and other physical attributes. Mostly what Tilden called physical condition.
  2. Technique: Swing mechanics, footwork, and associated elements.
  3. Game: Tactics, strategy, and fellow travelers. Similar to what Tilden called intelligence, with a bit of what he called experience.
  4. Competitiveness: The ability to sustain one's play and succeed in the face of various stressors.
  5. Confidence: The reliability of a player's judgement.

This model is an abstraction of the player. Each of these five contributors requires considerable discussion, and we will get to that momentarily, but given this taxonomy, a player can self-evaluate to get an estimate of overall player quality, giving each topic a total of 20 possible points. And similarly, a player can evaluate an opponent to get an understanding of the match up. All too often, we see players focus on one of these aspects at the expense of the others (not enough tent poles). More often than not, players will obsess on athleticism, or even a single aspect of athleticism. They will work out in the gym, trying to build more muscle that helps them lift weights but does little to help them with a racket. But teachers can also be guilty of obsessing on one aspect. Typically, when a teaching pro obsesses, it's on technique. We've seen lots of players who have reasonably good technique suffer because their teachers never get past the technique and onto the other 80% of play. So at first blush, our model for a player looks like this:

We'll discuss these contributors now, in the order in which they usually develop in a player.

Athleticism (0-20)
Athleticism is perhaps the most basic part of play, in that it's something that beginners bring to the game. They can develop athleticism as part of their progress, but they always bring something of athleticism into their play. We divide athleticism into nine components.

  1. Strength: The ability to do work with muscles. This is often called "static strength" in the functional training community.
  2. Size: The height, length, width, and sometimes the mass of a player.
  3. Speed: The ability to move fast over distance.
  4. Quickness: The ability to react to events in a very short time.
  5. Agility: The capability for rapid whole-body movement with change of velocity or direction in response to a stimulus. Some components of agility are typically called "dynamic strength" in the functional training community. In essence, dynamic strength is how many athletes achieve agility.
  6. Coordination: The ability to orchestrate muscles in unison. This includes hand/eye coordination.
  7. Flexibility: The ability to move joints with minimal effort.
  8. Balance: The ability to hold unnatural positions with minimal effort.
  9. Stamina: The ability to perform continuous work without exhaustion; basically, cardio-vascular fitness.

Different racket sports may favor one or another of these athletic components, but all of these contribute to every racket sport. For instance, table tennis favors quickness. The ball is moving at 100 MPH over a 9 foot table. That gives a total time of less than one-tenth of a second to react and take your shot. So quickness is essential. Squash, by contrast, favors stamina. Rallies are brutal and long. Professional squash tournaments often instrument players with heart rate monitors, and much like tennis tournaments display service speed for spectators, squash tournaments display heart rates. SquashMad.com reports that In a match between Ramy Ashour and Tarek Momen (two of the best players of the early 21st century) both players regularly peaked between 191 and 199 BPM during the match, and the average heart rate throughout the monitored time was 171. This would be life threatening for any normal person, but for top squash players it's all in a day's work.

Although basketball Coach John Wooden famously quipped that "you can't train height", most of the rest of this athleticism stuff--including size--can be trained. If you score yourself low on athleticism, you may be able to improve as a player by identifying which of these nine areas are weak, and working on those. To reiterated, one pattern you'll see in this discussion is to work on the weak stuff. If you're strong as a bull but can't touch your toes, maybe take a break from the weight room and spend some time in the yoga studio lengthening your muscles.

These criteria create an interesting fabric for comparing the athletic requirements for different racket sports. We've assigned seven racket sports requirement levels between 0 and 10 for each of the nine factors. The scores are based on our experience with the sports, and we think the scores are fair.

One thing to notice here is that in racket sports, agility is uniformly more important than static strength. And size is relatively unimportant, except for tennis. So what's up with that little quirk? The tennis serve is arguably the most difficult movement in all of racket sports, largely because it requires tremendous spin to be put on the ball in order to achieve any power...for most of us. Geometry just doesn't afford the opportunity to hit a flat ball with pace that both clears the net and stays in the service box. If we want to rely on gravity to keep the ball in the service box, that requires slowing the serve down. But if you can strike the ball at a height of 10 feet, the geometry changes. you can hit a flat serve that clears the net and does not rely on spin to bring it down inside the service box. This is why the current great servers, like Ivo Karlović (6'11"), John Isner (6'10"), Milos Raonic (6'5"), and Nick Kyrgios (6'4"), are all taller than 6'3", which is about the height one needs to be in order to strike the ball at or near 10 feet. To be sure, the great servers still put good spin on the ball. And shorter players like Roger Federer, who is "only" 6'1", have developed monumental serves. But height is a decided advantage on a tennis serve, and the taller a player is, the more margin for error they have due to the basic geometry of the game, and the more tools and variations they can put into their serve. The the basic mechanics changes to advantage taller players. Too much height may foreshorten a player's career through injuries and joint problems, but it gives a big edge, on-court. Incidentally, the real "minimum" contact height for a truly flat serve is about 9'3" rather than the nominal 10 foot. But that assumes you get everything perfect: you just clear the net (and at its lowest point), you hit just inside the service line, etc.. And if a 6'1" player strikes the ball at his or her highest point of reach, that results in a contact height of 9'3". Again, that's with everything perfect, hence the Federer serve at 6'1".  
    When we show this data in the table as a radar or "spiderweb" chart, some other interesting things pop out. 

    • While the athletic requirements for tennis are a bit more than for the other racket sports, the difference is contained mostly in the "size" component, which, as Coach Wooden conceded, is not something you can train.
    • Squash comes out next highest in athletic requirements, but the margin of difference is in stamina. Squash can be a grueling slog with 60 shot rallies being common in advanced play. Lots of people die on the squash court. An October 2008 story in Vanity Fair proclaimed squash "The Most Dangerous Game", and a number of academic studies have been done to understand why there are so many sudden deaths on the squash court. The reasons are pretty straight forward. Squash pushes the heart more than other sports. So squash is a very good sport for your health, so long as you don't die playing it.
    • Table tennis comes out low in athletic requirements, but this is largely due to the absence of a serious strength requirement. The racket is feather-light, at about 90 grams, and table tennis strongly favors flexibility, quickness and agility over strength. The flexibility requirement may not be clear...until you try to retrieve a ball that drops below the level of the table. Then the flexibility requirement becomes immediately obvious. And while size is a factor in table tennis, it is opposite of most racket sports: table tennis favors smaller players. It's harder to hit to the body of a smaller player. All of this explains why ten and twelve year olds can compete favorably in table tennis at the regional level. Tomokazu Harimoto, the Japanese ping pong prodigy, started playing ping pong when he was just 2, and was reaching the finals of major world level tournaments at 14 years old and about 130 pounds. No normal human could have any hope of swinging a tennis racket at two years old (note that Harimoto had to stand on a platform to play at that age in order to reach the table, but height adjustments for the youngest players are common in table tennis). A useful observation from this is that kids interested in racket sports at a very young age might start in table tennis years before they are ready to start in other racket sports. Developing the agility, muscle memory, and hand-eye coordination at a very young age may be a superb preparation for "heavier" racket sports later on, as well as giving them a game they can play throughout their lives, rain or shine. A fairly common sight at table tennis tournaments is an eight year old whose shoulders are at table level, competing against a wizened octogenarian, and having a great match. 
    • The chart also shows that table tennis has lower stamina requirements. We have two strong pieces of evidence to support that claim. Firstly, our own work with heart rate monitors indicates that table tennis seldom brings a player into what are traditionally called the "exercise zones". Typical heart rate levels are just below the exercise zones. The rapid-fire adjustments to the ball make it feel like you exert more energy than you actually do. If you can't get the ball in one step, it will be past you anyway. This leads to a different kind of workout. The other piece of evidence is in tournament structures. A typical table tennis tournament starts with a full round robin, and the results of the round robin are used for seeding in an elimination tournament. That's more than twice the match time than other sports.

    But let's take a little tangent on athleticism, and look to see how racket sports compare to other sports. We've averaged the values for all of the racket sports to get an aggregate profile, which we show in the chart below as the thick blue line. We've given our own estimates for the other sports, and again, we think these are fair. 

    There’s a lot in this chart. For one, football and basketball players are incredible athletes! We score basketball unusually high on the agility scale because it requires all of the same kinds of agility as football and racket sports, with the additional requirement for extreme vertical agility. And taking John Wooden’s “can’t train height” quip to heart, basketball players are very special athletes indeed. Which goes a long way toward justifying such high salaries for their performances. There are, in fact, certain similarities between basketball and enclosed court racket sports like paddleball and racquetball. In particular, court space is contested and both teams use their athleticism to take control of portions of that space, without too much bumping and pushing. To be sure, there is also a corresponding tactical issue of determining which portions of that space are worth the resource expenditure to control. A common pattern in a paddeball rally is for the striker to sweep into striking position in a way that moves the opponent out of coverage, and then shoot into the vacancy that was created. That pattern could have been pulled from the pages of a book on basketball. This pattern does not present itself in the net sports like tennis and padel where the net physically divides the court space and separates the teams.

    Another interesting point in the chart is the similarity in athletic profiles between football and racket sports, where the primary differences are in the unusual size and strength requirement. ts for football. From this, we would expect football players to spend a lot more time in the weight room than racket sports players, and for the most part, they do.

    We'd also like to draw attention to the powerlifting profile in the chart above (in mauve). It has a sort of keyhole shape. Basically, this means that it emphasizes a couple aspects of athleticism while others are not very important. In this case, size and strength are key. We would get a similar picture for bodybuilding or strongman/strongwoman athletes. Marathoners also have a keyhole profile. Speed, size, and stamina are key. But other athletic aspects are not so important. We know a math professor who says: "As a mathematician, I don't have the coordination or agility to chase and hit a ball. But also as a mathematician, I have an extremely high tolerance for pain, and so I run super-marathons." The takeaway point is that track and field athletes and even endurance athletes have a pretty narrow athletic profile. A perfectly balanced athletic profile looks like a circle (actually, a regular polygon) on the chart above. Football, basketball, and racket sports are all really well balanced. But football isn't a lifetime sport, while basketball and most racket sports are, and the ability to play a sport throughout your whole life has some value.

    Technique (0-20)
    Technique is the physical tool set that you use to play. It typically develops after athleticism, but usually before a player develops a game. Technique includes:

    1. Swing Mechanics: The ability to strike a ball (or a shuttlecock) effectively.
    2. Racket Preparation: The ability to maintain the racket in position to strike.
    3. Eye Control: The ability to focus on fast moving or subtle things.
    4. Footwork: The ability to move fluidly and efficiently.
    The specifics of all of these things vary by sport. For instance, a table tennis player may have great eye control on a small table and follow the ping pong ball skillfully, but have serious problems tracking a tennis ball across the 78 feet of the court from baseline to baseline. But conceptually it's the same thing. The difference, in this example, is most likely in the eye muscles which, like other muscles, tend to optimize themselves to the specific tasks they are forced to perform. A table tennis player with good eye control will tend to develop good eye control in tennis with continued play because the eyes adapt to the new task they are forced to perform. 

    Game (0-20)
    A player's game usually develops after he or she develops playing technique. They know their swing mechanics and their shots, and now they use those shots to produce points in rallies. A player's game includes:

    1. Tactics: A system for responding to situations on the court.
    2. Strategy: An overall approach aimed toward prevailing in the competition.

    Tactics and strategy are sometimes confused. To help understand the differences, tactics are what a Lieutenant uses to take a hill in battle, and strategy is what a Field Marshal uses to win a war. Charlie Brumfield is one of the best racquetball and paddleball players ever to step onto the court, and his game skill has allowed him to play at an extremely high level into his 70s. In a rally, Charlie has a tactical system that tells him what to do in each situation. When he moves into coverage, he anticipates two possibilities: the opponent's most likely shot, and the opposite shot. Charlie expects the most likely and prepares for the opposite. This makes it exceedingly difficult for the opponent to choose a winning shot. That's tactics. That's the Lieutenant. But Charlie is also a Field Marshal. His tactical plan services the strategic plan, which is typically to divide and conquer: to identify the part of the court that the opponent can defend as well as the part that the opponent cannot defend, and to grow the part that the opponent cannot defend with each successive shot in the rally until he can safely put the ball into that part of the court. That's strategy. That's the overall plan to prevail in the match. Both tactics and strategy can be learned on-court and off-court, but tactics probably favors on-court training whereas strategy probably favors training in the clubhouse. 

    Competitiveness (0-20)
    Some people are naturally competitive. Others aren't. There are players who have a great game, but can't pull out the big wins. Some players regularly make it to the semifinals and then wither. We suspect that a big part of that is that they have never trained for competitiveness. It can be trained, and it doesn't require Vince Lombardi's "winning isn't everything, it's the only thing" attitude. Competitiveness is two main things:

    1. Enduring Stress: Sustaining your ability to play in the face of stressors.
    2. Imposing Stress: Placing stresses on the opponent to impair his or her ability to play.

    Stressors can take many forms: injuries, weather, court surfaces, heckling, fear of an opponent, fear of critical situations, fear of crowds, fear of the referee, etc. Likewise, stressors can be imposed by a player to impair an opponent. Terry Gross did a famous interview of Andre Agassi in 2009, shortly after Agassi's autobiography Open was published. It's one of the best sports interviews ever conducted and deserves a listen by any serious racket sports player. Amongst other things, they discussed Agassi's last real professional match at the 2006 US Open against Marcos Baghdatis in the second round. Terry reminds Agassi that he was fortunate that Baghdatis sustained an injury in the fifth set. But then Agassi asserts that he "earned" that injury. That he and Baghdatis both played the most brutal form of tennis, one designed to take everything out of the opponent until they have no will to continue. That is just one way to impose stress on an opponent. And in that particular match, Baghdatis did much the same, moving Agassi left to right through long rallies that aggravated Agassi's existing back problems. After that match (which Agassi eventually won after more than four hours of play), both players required stretchers. An opponent can be stressed indirectly by "working" a referee. An opponent can be stressed by delays or by trash talk. There are lots of ways that one player can put a thumb on their side of the scale. But there are also lots of other ways that a player can put stresses on themselves: a "bad call" from an umpire or referee; a missed shot, especially a shot that just barely misses; an opponent's shot that shouldn't have worked but did, like one of those tennis shots that was supposed to be a deep drive but hit the tape and barely rolled over the net; an "act of nature", like a gust of wind that blows the ball; equipment failure; etc. In The Inner Game, Gallwey talks about a match he played against Jimmy Connors where he won the match, but Connors didn't concede. He even walked off, and somehow that caused the umpire to reverse the call, and when play continued, Gallwey fell apart and Connors went on to win the match. The ugly truth of racket sports is that there is a whole lot more that can go wrong than can go right. Anything that goes wrong can cause stress. And so a competitive player learns how to cope with stress. Perhaps another way to describe competitiveness is using all of your skills to make as much as possible go wrong for the opponent and as little as possible go wrong for you.

    Confidence (0-20)
    Before Super Bowl I, Fred Williamson of the Kansas City Chiefs (nicknamed "The Hammer") captured national headlines by saying that he would knock the Green Bay Packers' starting receivers, Carroll Dale and Boyd Dowler, out of the game. "Two hammers to Dowler, one to Dale should be enough," he boasted. That's not confidence, at least not as we mean it. That's cockiness. And "The Hammer" got hammered, by Packers' running back Donnie Anderson in the fourth quarter and "The Hammer" himself was knocked out of the game. Confidence as we mean it here has nothing of that. It's more akin to what statisticians call a "level of confidence", or what practitioners of the neurobiomechanical model call "neuro-technique". Confidence is to the brain what swing mechanics and footwork are to the body. The point of swing mechanics and footwork are to be efficient, reliable, and repeatable ways to move the body through a match. Confidence is about efficient, reliable, repeatable ways to move the mind through a match. It's about the reliability of one's judgement. And a player's judgement lives on long after the player is gone, which is why we still read Bill Tilden 65 years after his death. Confidence has four basic components:

    1. Self-Esteem: An understanding of what you can do (in the context of your opponent's play).
    2. Humility: An understanding of your limitations (in the context of your opponent's play).
    3. Actualization: The ability to see your best shot before you take it, and then make the shot. Likewise, the ability to anticipate your opponent's play.
    4. Doubles Confidence: The ability to support a doubles partner through periods of stress, panic, and/or withdrawal.

    Some players just seem to consistently be at the right place at the right time and make the right shot. Squash player Rami Ashur is an example. Roger Federer in tennis. Lee Chong Wei in badminton, Ma Lin in table tennis. One of our favorites, one-wall paddleball and racquetball player Robert Sostre. These players tend to have long careers because they can compete at the highest level even as their athleticism tarnishes. They know with high certainty what they can do, what they are going to do, and then they do it. An example of confidence is: "I am confident that I can make this crucial shot effectively into my opponent's backhand because my pace is sufficient, and I am also confident that I cannot make this crucial shot effectively into my opponent's forehand because my deception is not good enough and my opponent is too quick". Actualization is really the embodiment of confidence. The actualization part is seeing this all play out in your head beforehand. A player who is genuinely confident will see the shot before taking it, and just know that he or she will make the shot so visualized, like a Larry Bird jumper. Confidence grows naturally with experience, but there are also ways to train confidence.

    We also include here a special kind of confidence we call "doubles confidence". Most racket sports have a doubles game. Typically, doubles games are more strategic than singles. An important aspect of doubles that simply doesn't apply to singles is partner management. There's a lot of debate about the so called "hot hand theory", which states that players have periods when they are hot, and other periods when they are cold. Some analysis suggests that this is a "hot hand fallacy"; that there really are not hot and cold periods. It's hard to look at the history of an athlete like Tiger Woods and not identify clear hot and cold periods. Maybe in blackjack there are no real hot and cold periods, but in sports there are. There are hitting streaks in baseball. And pass receivers like Jerry Rice and Lynn Swann in football have had streaks where they couldn't drop a ball. they just get in a zone. And there are cold streaks too. Slugger Dave "Kong" Kingman could hit three home runs one game, and then strike out four times the next. Hot and cold streaks arise mostly because of mental factors, and mental factors are a big part of racket sports. A player can get intimidated by an opponent and become hesitant and make multiple consecutive unforced errors. In doubles a player can also become intimidated by a partner and make multiple consecutive unforced errors. This happens, for instance, if a player makes a mistake or goes into a slump and the partner turns a cold shoulder as if to say: "okay, you're on your own to make your way out of this". The partner under stress in that situation begins to buckle, and make obvious mental errors. On the other hand, a good doubles player knows how to share confidence with a partner in a time of need; how to bring a partner out of a slump and how to keep a partner from getting tight in crucial situations. This is "doubles confidence" and it is a big deal in doubles play. If your racket sport is pickleball, doubles confidence is probably the most important part of the game. Two octogenarians with shaky knees and bad eyes can make a formidable pickleball team if they are each superb at taking care of their partners.

    In order to better understand confidence we begin with an example from weightlifting. No one is more focused than a great lifter doing a great lift. Here is a video of Eddie "The Beast" Hall smashing the deadlift world's record.

    It's a unique lift because he isn't just pushing the record by a feather or two. He's crushing the record by about 70 pounds (an 1100 pound deadlift). And so Eddie spends about a minute preparing for his lift, building the confidence that he can actually make it, and then about 5 seconds on the actual lift. To be sure, Eddie Hall truly is a beast, and a massive one, and his deadlift technique is top drawer, but that lift is as much about confidence as it is about beast. Racket sports athletes don't really have that degree of intensity. Jimmy Connors was intense; Serena Williams is intense; in table tennis, Ma Lin was intense. But Eddie Hall is off the charts. As he steps to the bar, no one has ever succeeded at this lift. No one has even tried seriously. He's failed previously at lesser lifts. If he doesn't believe he can make the lift he surely won't, so how can he bring himself to believe he can succeed? The one thing he has is all the time he needs to psyche himself up for it. Racket sports players don't get that even on the serve. But this is a good opportunity to analyze confidence from a game theoretic perspective, and thereby get an understanding of the similarities and differences between weightlifting confidence and racket sports confidence. 

    When we play a game, we have opportunities to make choices, or "moves" if you prefer. Associated with each move is a payoff. But when we choose a move, that choice is based not on the actual payoff, but on our perceived estimate of the actual payoff. In some games, there's no difference between the actual an estimated payoff, but those games aren't very interesting. These are really just force-on-force encounters and the stronger player wins. That's how casinos make money. They set up the rules of the game so that the house is slightly stronger than the player. The interesting games are ones where the estimated payoff may be different from the actual payoff. 

    In the games we are talking about here, the player's confidence is really his or her estimated payoff function. If their estimate of the payoff is in error, there is likely to be a cost in performance as they play the game. Now let's apply this to the weightlifting situation. And let's assume that if the lifter doesn't believe they can make the lift, they won't. That's a simplification, but it's almost certainly the case, especially as the lifts we are concerned with are the pinnacle lifts at the lifter's maximum potential. But if they can't make the lift, then confidence isn't going to help. They simply can't do it. Consider the following chart, which uses notional values. If the lifter makes the lift, he wins the competition's prize money. If he doesn't make the lift, he wins nothing:

    The main diagonal in the chart shows the outcome if the estimated payoff had no error and was exactly the actual payoff (best case scenario). The other elements in the column show actual results in the face of estimation error; underconfidence or overconfidence. The cost of estimation error can be found by subtracting an off diagonal element from the diagonal element in the same column. This shows that the cost of underconfidence is in the first colulm. The lifter could have won the competition if he only believed. But since he didn't believe he goes home without the prize money. And the cost of overconfidence is in the second column, and, well, there is none! The lifter would have lost anyway. There is something profound in that, so let's repeat it. In the weightlifting scenario, there is no cost paid for overconfidence. That's why the walls of weight rooms are papered with motivational slogans like: "If you're afraid to be strong then you deserve to be weak". This kind of conditioning will almost certainly lead to overconfidence in the weightlifter. But there is no price to be paid for overconfidence, while there is a hefty price to be paid for underconfidence. So weightlifters should be as overconfident as they can to eliminate the possibility of an underconfidence error. That's a slight simplification. Overconfidence could get a weightlifter injured. Eddie Hall gave himself a concussion with his lift, he had a brain bleed and he couldn't recognize his wife or children for a month, so in that sense he might have been too confident, but in terms of the payoff for the competition, the model is pretty accurate. 

    Now let's do the same exercise for racket sports. Again, we'll simplify the real world a bit (that's what models do), but we'll do so without losing any of the essential qualities of the real world (that's what models are supposed to do). Here we will assume that the overconfident player attempts to hit an attacking winner, and the underconfident player attempts to hit a neutral shot that leaves the rally in the same basic position it was in before the shot (e.g. a "pusher" in tennis). Further, we'll assume that the player can always make the neutral shot, and there are different columns for whether the player can hit the winner. And finally, we will assume that if the attempted winner misses, the opponent wins the rally (skip ball in racquetball or paddleball, or in the net in net sports).

    As with the weightlifter, there is a cost for underconfidence. We could have hit the winner, but we didn't. But there is also a cost for overconfidence. We hit the ball into the net and gave the opponent the point by being too aggressive when we could have kept the rally going. To summarize, for weightlifters the best confidence estimator is maximum overconfidence, but for racket sports players, the best confidence estimator is identical to the actual payoff function. Two players can come onto the court with identical skills but if one is better at estimating his own ability to hit shots, that player will win in a blowout. One consequence of all of this is that players who go into the weight room to get stronger sometimes come out with what we'll call "weightlifter confidence" and that serves them poorly on the court. So a big goal for us will be to learn how to internalize a payoff function that matches the real world. 

    Player Model Summary

    We now have a more complete model of the player.

    The criteria are now fine enough, and concrete enough, that a player should be able to self-evaluate him or herself and also to evaluate an opponent. The big questions are how to train to improve one's score as a player, and how to use your evaluation of your opponent to gain an advantage. That is the subject of the next section. 

    Now that we know where we stand as a player, the goal is improvement. One straight forward observation is that if you already score 20 for an aspect, no improvement is possible, so it's a fool's errand to focus your training there. You've already topped out. And if you currently score a 19, there's very little room for improvement, so there are few ways to improve and improvement will be difficult. If you currently score 18, there is a little more room for improvement, so it's easier than if you score 19, but it's still very difficult. Moreover, on average, the better you are at an aspect, the harder it is to improve. We've drawn a notional chart to indicate the level of effort required to increase your score when starting at a given level.

    The obvious conclusion from this chart is that to train efficiently, you want to work to improve the areas where you are weakest because the effort is lower for a given level of improvement. If you score a 5 on fitness and a 15 on confidence, it's a whole lot easier to bump the fitness up to 6 than it is to bump the confidence up to 16. Unfortunately, lots of players train their strengths. They've had success there in the past, which is how they got strong in that area, and so they go back to the hole. Then they reach plateaus that they can't get out of. Likewise, we have seen coaches who don't want to move on to game aspects until the player's technique is a perfect 20 because they are really good at training a player's technique. Again, they go to the place where they've had success. But getting to the perfect 20 is almost impossible. And so the player languishes in intermediate levels with a very nice swing while easy improvement is available in other aspects of the player. The player becomes frustrated losing to players with far weaker technique, and doesn't understand why, because they haven't learned that game, competitiveness, and confidence need to be included in the evaluation of their opponent. Their model of a player is deficient. And because it's a predictive model rather than a causal one (e.g. "get your swing mechanics right and you'll win, trust me") they don't have the perspective to improve the model. They don't know why their training isn't working. Eventually, they leave the sport disappointed.

    All five aspects of a player: athleticism, technique, game, competitiveness, and confidence can be trained and improved. In fact, there is at least some element of each that can be trained on-court, and some element of each that can be trained off-court. Off-court training is particularly useful to those of us who don't live at the club. Off-court training allows us to make improvements while at home or work. On-court training is usually more fun. Not always, but usually. Some on-court drills are brutal.

    We will examine two different approaches to improvement. The two approaches can, and should be used together.

    1. Focused Training: Isolate an aspect and focus on improving it.
    2. Cross-Training: Identify the stronger aspects and use those as tools to improve the weaker aspects.

    Focused training is good, and we will deal with that first, but cross-training is the most important way you can train, and we will cover that momentarily.

    Focused Training
    In focused training, we isolate an aspect of play and work diligently to improve it. Usually, this means looking deeper into that aspect to find the specific component where the weakness is, and training that. 

    The focus of our overall discussion is not gym workouts. But let's say your self-evaluation score on athleticism is low. We need to provide some course for improvement. But "athleticism" is too broad to train for improvement. Training to be more athletic is like studying to be smarter. The categories are too broad and without refinement, failure is likely. So you identify which parts of your athleticism are wanting: strength, size, speed, quickness, direction change, coordination, flexibility, balance, or stamina. Each of these is specific enough to improve. For each part of each aspect, we give some general ideas for improvement. These are not exhaustive, but provide a reasonable outline for a training course to improvement, and given these examples, it's usually possible to find other means of focused training to improve that aspect.
    • Strength: Strength training can be done on-court or off-court. On-court, you can wear a weight vest, preferably a light one. You also might weight the racket or wear ankle or wrist weights, but we recommend against that as this can lead to injury, but some players have had success that way. However, off-court, any of these will work. You can wear a (possibly heavier) weight vest at work and train your strength throughout the day. Wear ankle or wrist weights on your evening stroll. Even one pound "knuckle weights" will make your walk more challenging. You can workout in the weight room if that suits you. And if you want to reach your maximum potential of athleticism, you'll have to. However, you should be aware that few weightlifters or bodybuilders end up as good racket sports players. They tend to build bulk that gets in the way of their own movement. The body's muscles are largely divided into fast twitch and slow twitch muscles. The slow twitch muscle groups tend to be larger, and so these are the ones "physique athletes" usually try to build. Racket sports rely much more on the smaller fast twitch muscles. Big muscles often just get in the way, and so racket sports athletes tend to do best in the weight room when they introduce "balance challenges" into the exercise. Here, for example, is a photo of Chinese table tennis team player Liu Guiliang doing dumbbell raises while standing on physio-balloons to challenge his balance in the lift. The balance challenge makes the exercise more difficult. As a consequence, he can use smaller weights which reduces the risk of injury. In addition, the balance challenge engages the fast twitch muscles from the balls of his feet up to his wrists. This kind of strength training is one of the tools that got Liu Guiliang into the world's top 4 in his sport.

    Bosu Balls are also excellent for engaging the fast twitch muscles, and can be substituted for the balance balloons Liu is using in the photo. The Bosu Ball is like half of a physioball (a hemisphere). It can be turned so that either the flat side or the round side is down. Round side down tends to be more challenging. For most racket sports strength training, round side up works fine, but if you want the extra challenge, turn the round side down and balance on the flat side. Some athletes actually use a physio ball (full-round ball, aka "Swiss ball") in this way. For instance, they step on the physio ball with a weight bar and do curls or even squats. Here's a YouTube video to prove it.

    YouTube Video

    This is quite dangerous, and we don't recommend it, even if you see others in the gym trying it out. It's a good way to get hurt. We only know one good racket athlete who has successfully trained this way, and not since the concussion. There are lots of ways to use a physio ball as a balance challenge for weightlifting: sitting on it is okay. Doing squats with the physio ball between you and the wall is good. Standing on it is usually not smart. 

    Bicycles are also a useful way to build strength. You can ride hills, or, if you are stuck with a stationary bike, crank it up to a high level and do interval training on it. Some gyms have stationary bikes that are specially cranked down to make them more resistant for strength training. Ride for two to three minutes on a setting that causes your quads to burn, rest for a few minutes, and ride for another couple of minutes. 

    To train specifically for wrist and forearm strength, gyro balls like the PowerBall are really good. 

    We actually prefer the gyro balls with the handle, but those don't seem to be available anymore. With the gyro ball, you start the inner ball to spin and then use your wrist to keep the ball spinning. you can get these up to a pretty good speed (thousands of RPMs). You may have heard tennis or table tennis players talk about a "heavy ball" as one with a lot of spin. As you spin the gyro ball faster and faster you will know what they mean. At rest, the gyro ball probably weights about 8 oz. But as you spin it up faster and faster, it will feel like it weighs five pounds or more. And you will get a serious forearm workout by keeping the ball spinning. But don't overwork the gyro ball. Too much gyro can aggravate tennis elbow. The more expensive PowerBall models have built-in sensors that will give you the gyro ball's RPMs. That's a nice feature, but you pay a lot for the sensor. Here is a reasonably good online video tutorial to help you figure out how to use the gyro ball.

    To be sure, the functional training community of sports physiologists has a lot to say about strength training for sports. We highly recommend Functional Training for Athletes at All Levels by James C. Radcliffe who is the head strength and conditioning coach at the University of Oregon. Radcliffe has trained many of the athletes in the University of Oregon's exceptional NCAA and intercollegiate sports programs, as well as authoring a number of scholarly papers on sports physiology. We note that the book focuses on the athletic requirements of the big NCAA team sports like football and basketball, but most of what he says can be applied to racket sports as well. Radcliffe gives more than 100 exercises tailored to the different aspects of (functional) training. Most of Radcliffe's exercises don't require any sundries like a weight bar. He also discusses warmup and gives a good discussion of dynamic warmup. 
    • Size: In general, size in court sports is a bad thing. To be sure, tennis players taller than 6'3", like John Isner, have some advantage in that they can hit a flat serve that doesn't require gravity to pull it down. But taller players also have an array of injury problems. In table tennis, small is usually the rule. In fact it can be difficult to find table tennis shoes larger than a US 11.
    The more important issue of size is weight, and most amateur players are heavier than ideal. One way to reduce the weight is to play more, particularly in racket sports like squash or paddleball that burn a lot of calories and increase metabolism. We recently had the pleasure of watching a slightly plump former racquetball pro try his hand at squash. After he finished his match and slid his body down the outside of the glass and into a lump on the floor he looked around the squash club and noted: "There. Are. No. Fat. People. Here!" This is actually a pretty good yardstick to determine whether trying to cross-train in a different racket sport will help. If all of the people at the club where they play that sport are lean and wiry, and you're trying to lose some weight, maybe try a game or two.
    But just as diet can't compensate for lack of exercise, exercise does not make up for bad diet. There are obviously a lot of weight loss diet plans out there, and if anyone knew what the answer was, everyone would be svelteProbably the most important consensus point is to maintain weight by building a diet that emphasizes nutrient-dense plant products, and avoids "ultra-processed foods". It would be good to know what we mean by "ultra-processed foods", and this can be confusing. But in 2016, Carlos Monteiro, a Professor of Nutrition and Public Health at the University of Sao Paulo and a consultant to the World Health Organization (WHO), published a very useful way of classifying foods in the journal World Nutrition. The system is called NOVA, which is an accepted standard of the WHO. NOVA is a name and not an acronym, and so it can't be expanded. But it gives four levels of food types (which we quote, abridged, here directly from Dr. Monteiro's report):

    • Group 1:  Unprocessed or minimally processed foods
    Unprocessed (or natural) foods are edible parts of plants (seeds, fruits, leaves, stems, roots) or of animals (muscle, offal, eggs, milk), and also fungi, algae and water, after separation from nature. Minimally processed foods are natural foods altered by processes such as removal of inedible or unwanted parts, drying, crushing, grinding, fractioning, filtering, roasting, boiling, pasteurization, refrigeration, freezing, placing in containers, vacuum packaging, or nonalcoholic fermentation. None of these processes adds substances such as salt, sugar, oils or fats to the original food.

    • Group 2:  Processed culinary ingredients
    These are substances obtained directly from group 1 foods or from nature by processes such as pressing, refining, grinding, milling, and spray drying. The purpose of processing here is to make products used in home and restaurant kitchens to prepare, season and cook group 1 foods and to make with them varied and enjoyable hand-made dishes, soups and broths, breads, preserves, salads, drinks, desserts and other culinary preparations. Group 2 items are rarely consumed in the absence of group 1 foods.

    • Group 3 Processed foods
    These are relatively simple products made by adding sugar, oil, salt or other group 2 substances to group 1 foods. Most processed foods have two or three ingredients. Processes include various preservation or cooking methods, and, in the case of breads and cheese, non-alcoholic fermentation. The main purpose of the manufacture of processed foods is to increase the durability of group 1 foods, or to modify or enhance their sensory qualities.

    • Group 4 Ultra-processed food and drink products 
    These are industrial formulations typically with five or more and usually many ingredients. Such ingredients often include those also used in processed foods, such as sugar, oils, fats, salt, anti-oxidants, stabilizers, and preservatives. Ingredients only found in ultra-processed products include substances not commonly used in culinary preparations, and additives whose purpose is to imitate sensory qualities of group 1 foods or of culinary preparations of these foods, or to disguise undesirable sensory qualities of the final product. Group 1 foods are a small proportion of or are even absent from ultra-processed products. Substances only found in ultra-processed products include some directly extracted from foods, such as casein, lactose, whey, and gluten, and some derived from further processing of food constituents, such as hydrogenated or interesterified oils, hydrolyzed proteins, soy protein isolate, maltodextrin, invert sugar and high fructose corn syrup. Classes of additive only found in ultra-processed products include dyes and other colors, color stabilizers, flavors, flavor enhancers, non-sugar sweeteners, and processing aids such as carbonating, firming, bulking and anti-bulking, de-foaming, anti-caking and glazing agents, emulsifiers, sequestrants and humectants. Several industrial processes with no domestic equivalents are used in the manufacture of ultra-processed products, such as extrusion and moulding, and pre-processing for frying. The main purpose of industrial ultra-processing is to create products that are ready to eat, to drink or to heat, liable to replace both unprocessed or minimally processed foods that are naturally ready to consume, such as fruits and nuts, milk and water, and freshly prepared drinks, dishes, desserts and meals. Examples of typical ultra-processed products are: carbonated drinks; sweet or savory packaged snacks; ice-cream, chocolate, candies (confectionery); mass-produced packaged breads and buns; margarines and spreads; cookies (biscuits), pastries, cakes, and cake mixes; breakfast ‘cereals’, ‘cereal’ and ‘energy’ bars; ‘energy’ drinks; milk drinks, ‘fruit’ yogurts and ‘fruit’ drinks; cocoa drinks; meat and chicken extracts and ‘instant’ sauces; infant formulas, follow-on milks, other baby products; ‘health’ and ‘slimming’ products such as powdered or ‘fortified’ meal and dish substitutes; and many ready to heat products including pre-prepared pies and pasta and pizza dishes; poultry and fish ‘nuggets’ and ‘sticks’, sausages, burgers, hot dogs, and other reconstituted meat products, and powdered and packaged ‘instant’ soups, noodles and desserts.

    The consensus agreement of current dietary studies for health and weight control is to build as much of the diet as possible around group 1, and avoid group 4. Montiero's report provides much more complete descriptions and explanations if the reader wants a deeper understanding. Something to be aware of is that those recovery powders and protein shakes and interworkout drinks that people suck down in the gym as a part of their health program are all in group 4, which we should stay away from if we want to maintain good health. The best plan is probably not to eliminate them completely, but to make sure that they are not a significant fraction of calorie consumption.

    NOVA can be thought of as a tool for building or understanding diet plans. For instance:

    • Bright Line Diet: Group 1-3, except with no sugar or flour of any kind (basically, the Bright Line Diet says that you need to put up hard prohibitions of certain foods, most notably sugar and flour).
    • Paleo: Restrict consumption to Group 1 foods.
    • Vegan: Groups 1-3, except with a bright line barrier barring the animal foods from group 1.
    • Intermittent Fasting: Allow any group during feeding periods and none of the groups during fasting periods.

    Racquetball hall of famer Shannon Wright is a strong proponent of Dr. Joel Fuhrman's 
    Eat to Live diet. Shannon is also a PhD/MD and knows her way around the human physiology, especially racket player human physiology. So Eat to Live might be a good place to start. Eat to Live is almost like vegan, except with dotted line barriers to the animal products rather than a bright line. Similarly, dotted line barriers to group 4. Dr. Fuhrman says that these foods can be eaten, but they should be treated like one would treat a recreational drug. To be clear, there is a certain amount of controversy about Dr. Fuhrman's plan, and he has been accused of over-interpreting his own research. But his claims, in broad strokes, are consistent with other independent research and so the plan is a good way to orient. Dr. Furhman acknowledges that strict adherence to his diet provides insufficient calories for athletes, but also notes that it can be adjusted to the activity level of your life.

    Eat to Live is based on nutrient-dense foods from group 1, mostly dark-colored vegetables and fruits, and to a lesser extent, the whole grains in group 1. Eat to Live discourages processing, and so groups 2-3 are sketchy, and group 4 is strongly discouraged. Fuhrman rates foods with an Aggregate Nutrient Density Index (ANDI value), where kale has the highest rating at 1000 and Coca-Cola has the lowest at 1. "Density" implies that you divide by something, like "pounds/cubic foot" or something similar. In the ANDI system, calories are in the denominator. So the vision is to take the aggregate value of thousands of nutrients and micro-nutrients and divide by the calories it costs you to get that nutrition. Fuhrman gives a page of ANDI values for common foods on his web site, and these might be good to look at even if you're already at a good weight. There is also the Eat Right America Nutritarian Handbook that gives ANDI values for thousands of different foods. This is different than a calorie counter because calories only consider one nutritional aspect (i.e. the potential for the food to make you fatter). We have certain issues with the way ANDI is calculated. Principally, it is supposed to be a density index, but that's not really how it's calculated, at least not consistently. A true density index would not depend on food quantity. Fava beans would have one nutrient density value, whether for an ounce or a ton. Some foods are given an ANDI value like this while others are per ounce or per cup, etc. This doesn't make sense and suggests that when the guides were put together, the ideas were stretched and squeezed in inconsistent ways. There is also Fuhrman's unsettling claim that many (thousands?) of micro-nutrients haven't even been identified yet. And some of the value choices appear questionable. For instance, corn on the cob rates as much better for you than olive oil. Most nutritionists would quarrel with that choice. And fish rates fairly low in Fuhrman's plan. This runs counter to current research. For instance, researchers from the Harvard School of Public Health and the University of Washington reported in 2013 in the peer-reviewed Annals of Internal Medicine that those with higher levels of fish in their diet showed a 35% lower risk of dying of heart disease compared with those with lower levels of fish in their diets. Because ANDI values punish food processing, fermented foods also tend to suffer in the rating. Kelly Bilodeau, Executive Editor of the Harvard Women's Health Watch writes in Fermented foods for better gut health (not peer-reviewed) that naturally fermented foods like kimchi and other pickles contribute to gut health and prevent neuro-degeneration. So to be sure, there are various points of contention with the ANDI scores. However, as a general guide as to whether a candidate food is worth the calories, the ANDI score is a good first check, and is particularly useful for comparing potential substitutes. We have lots of different choices for rice: white, basmati, jasmine, brown, red, wild rice, and black rice. Does it matter which you choose? Yes. There is a wide range. Black rice is much more nutritious than white rice. The others fall in the middle. That's good to know when you are planning a meal.  Again, Eat to Live and the ANDI values are only a starting place. Athletes have had success maintaining size with this and other plans.
    • Speed: Speed can be trained. We are not fond of wind sprints--especially for older players--due to their impact on the knees. We'd like to save as much of our lifetime knee impact budget as possible for the court. But one of the best ways to improve speed is with a bicycle. If you've got a bike that has a rack and pannier bags, use it when you need to get a few things at the store. Use it to go to the library. If work is within range, ride to work. Same for the gym. Bike to your workout and you get a warmup in without costing you any time. Jump ropes, particularly speed ropes are another great way to train speed. Buddy Lee trains Olympic athletes at the Olympic Training Center. He was an Olympic wrestler himself and has developed some impressive routines. Most of these are not specific to a particular sport. He also has created some great speed ropes. He has a lot of information on his web site, where he also sells his speed ropes. They are also available on Amazon.com. Here's a video Buddy uses to introduce the rope to potential trainees. It's quite impressive and deserves a view.

    But there are some important caveats about jumping rope. If done correctly, on the correct surface, jumping rope is low impact. Please do not jump rope on concrete or a hard court. At least not for more than a minute or two. We see jump ropes in cardio-tennis sessions on the hard court. Be very careful. If you play squash, badminton, or indoor racquetball on the other hand, a well-sprung hardwood floor like those of high quality indoor courts is ideal for jumping rope. Look for a sprung floor (like a good hardwood floor), rather than a padded floor (like a carpet). The surface needs to bounce back. If you don't have a good sprung floor available, a nice choice is a sheet of plywood over a dense carpet. This has the cushion of the carpet and the spring of the wood. The plywood can be stood up against a wall when not in use. The plywood sheet can also serve other functions. For instance, you can install some flush anchors on it to tie resistance bands to, and use it for resistance band squats. 

    Plyometric training is another way to improve speed. Plyometrics typically involve an aerobic step or similar step. You can do slalom lunges across the step, or similar exercises. The benefit is similar to the jump rope. We find it more boring than a jump rope but if it works for you it's probably good. Similar caveats as with the jump rope. Do your work on a surface with some give. Also, a lot of plyometric routines are based on box jumps. Box jumps are great, but you have to be very careful. If you are a bit too ambitious, you can cause yourself fairly severe injuries. 
    • Quickness: The best way we've found to train quickness is to play table tennis. If your main racket sport is something else, training with table tennis will really work on your quickness. Jumping a speed rope will also train quickness, but as before, try not to jump rope on concrete or other similarly hard surfaces. You can also drill, on-court, to train quickness. If you are a racquetball or paddleball player, you can stand about five feet from the front wall and do continuous reaction volleys with yourself. This will feel somewhat like a boxing workout on a speedbag, and your quickness will improve within a few workouts. If you play tennis, you can set the ball machine near the service line and assume a position at the net and practice volleys. Pretty much anything on a badminton court will enhance your quickness. And while the fast twitch muscles are the foot soldiers of quickness, the eyes and ears are the forward deployed commandos. Regardless of how you train, to become quicker, you need to learn to see and hear the play. You need to learn to identify and monitor the cues. In enclosed court sports like racquetball, one training method is to go onto a dark court and do your drills. You have to work harder to see the ball, and as a consequence, you place more emphasis on hearing the ball off the racket and the wall. But you can also train your eyes and ears by simply being more deliberate. On your evening walk, look and listen carefully. Try to pinpoint the bird you hear, and try to see any changes from your last walk. Similar exercises work on-court. Some coaches teach that your on-court warmup is not for your muscles (those should have been warmed up already, maybe on an elliptical trainer or stationary bike), but for your eyes. Start short and slow. Watch the ball carefully, and then gradually move back, hitting longer and finally faster balls, maintaining the same focus you had for the short, slow balls. 
    • Agility: Experts are still trying to come to agreement on what exactly agility is. And without a definition of the thing, it's really hard to put together a training program. Jeremy Sheppard and Warren Young proposed a good working definition in a 2006 paper in the Journal of Sports Sciences, a scholarly journal for sports physiologists. They say that agility is "a rapid whole-body movement with change of velocity or direction in response to a stimulus”. The change-of-direction part is key not just to agility, but to athleticism as a whole. Sheppard and Young go on to note that "Agility has relationships with trainable physical qualities such as strength, power and technique, as well as cognitive components such as visual-scanning techniques, visual-scanning speed and anticipation." This means that your off-court agility training can include training your eyes or watching video of play to develop better anticipation.
    In his interview with Terry Gross, Andre Agassi confided that one of the moments that changed his thinking about tennis occurred when he met Gil Reyes, his trainer for much of his career. Gil asked Andre about his workout, and Andre mentioned that he ran five miles every morning. Gil observed that tennis doesn't require you to run great distances, it requires you to run five steps, change direction, and run five steps the other way. Agassi's said that this was the smartest thing anyone had ever said to him about tennis. His realization of the importance of direction change completely altered the way he trained. Gil Reyes' main point--that changing directions is key to athleticism and player performance--also holds for other racket sports. And the fact that it took Gil Reyes for the hard-training Agassi to realize this suggests that elastic potential is often overlooked by players and coaches as a significant component of their athleticism. But let's examine the impact on your game. Say that the ball is moving at 60 MPH. That's probably a good round number for most racket sports. If you can improve the time it takes you to change directions by just 1/20 of a second, that makes a difference of about 5 feet. That's a full step on the fly. How many balls do you miss getting to by less than a full step on the fly, and how would it improve your play if you could reach all of those? What we are really talking about here is "elastic potential", which is an umbrella for springiness, bounciness, and the ability to quickly pivot and change directions. Skier drills on the jump rope are a good tool to train elastic potential (see the Buddy Lee video above for an example). But slide boards may be the best tool available to most players to train for this. However, many gyms don't have slide boards and so they can be dismissed by their absence. They are also not very expensive and most players can afford one of their own or convince their club that it's worth the investment. Most slide boards roll up for storage in small places.

    Here's a good video of speed skater Kevin Jagger training on a slide board. The video is in two parts, part one showing Kevin in his first days with the slide board and part two showing Kevin working the slide board a year later. The takeaway from the video is that on the bounce, where Kevin's foot pushes off the chock at the end of the slide board, he's cut the bounce time in half with that year of slide board workout. He goes from about two seconds to about one second on each bounce. So he's gained a second. In one second, a ball moving at 60 MPH travels more than the length of a whole tennis court. Of course the slide board isolates this kind of activity. Working on the slide board would be like playing tennis on glare ice while wearing brand new leather soled shoes. We don't do that. The actual gain is less than that full second, but it is still substantial, and probably exceeds the 1/20 sec. bogey. 

    We've mentioned that bouncing or "springiness" is a big part of elastic potential. You can see that springiness in Kevin Jagger's video. And the slide boards and jump ropes are really training springiness more than anything else. Here's a good experiment that's changed the way a lot of racket athletes train and play. In fact, it's changed the way lots of athletes in other sports train and play as well: Hold your hand out over a desk or table. Keeping your arm firm and just using your open hand, hit the table as hard as you can. Listen to the sound. Now, holding your hand out over the table as before, use your opposite hand to pull the fingers of the suspended hand up, like pulling back a spring or pulling a bowstring. Now release the bowstring and let the open hand that was sprung back hit the table. Listen to that sound. If you're like most people, that sound was much more powerful than the first. That's the difference elastic potential and strength. Elastic potential is explosive power. As another illustrative example, consider an offensive lineman in football. When the ball snaps and they start to block the opposing defensive lineman, they lunge into action. That's their elastic potential. After that, they push the defensive lineman back with strength. But that first explosive lunge is much more powerful than the rest of the block. But it's also much more important. In that situation, the offensive lineman (and the defensive lineman too for that matter) elastic potential is far more important than raw strength. This is one of the big "lightbulb moments" in what's known as "functional training", which is training for athleticism rather than training for physique. Most athletes train more for elastic potential.

    Slide boards were designed to train speed skaters. But they've been a secret training ingredient of some of the most successful racket sports players (we were introduced to them by Chris Crowther, former top 4 racquetball player). Basically, you put some silk booties over your shoes and slide back and forth on the slippery slide board surface. There are chocks at each end so that you don't slide off, and when you hit the chock, you change directions and go the other way. The caveats for slide boards are a different than for jump rope. Slide boards are very low impact since you are constantly sliding. And you need a smooth hard surface to slide on. So concrete is a very good surface for the slide board.     

    Because of the nature of speed skating, where the change in direction generates the thrust, skaters have to kick off from side-to-side and they use a lot quadriceps and calf muscles. We have better traction in racket sports. So we can get more of a spring from our shin muscles. In fact, these (principally the tibialis anterior that runs vertically just to the outside of the shinbone), are key to a racket athlete's bounce. The reason goes back to the goal of reducing the amount of time needed for the direction change. It takes time to coil up and uncoil the kinetic spring. It takes time to bend the knee, and it takes time to unbend it. And if the bend continues to the waist, more time. The more you can control the bounce with the spring of your ankle, near to the point of contact with the ground, the less time spent to coil and uncoil; the less time it takes to change direction. We still need the quads and the glutes that the slide board trains, but we need the shin muscles just as much or maybe more. These are the main bounce muscles. Some plyometric exercises are very good for these bounce muscles. Low box jumps are probably the best example. These can be done with an aerobic step. Typical aerobic steps come with some blocks that can be installed at either end, and this allows the height of the step to be varied. Use those to raise your bed or something. You won't need them here. The kind of jumps we want to do are just fine with the lowest step (no blocks on either side), because we are not interested in the whole jump, just the bounce; just the shin. In fact, a plain old board will do if you don't have an aerobic step. To do these jumps, stand in front of the step and using your shin muscles, bounce up onto the step and then bounce back down without touching your heels to the ground. Repeat. You are not bending your knees in this exercise. You don't want to lock them out, but you don't want to bend them either. It's not about the big muscles of the thighs. It's about training the shin muscles to bounce. When you pivot and turn on the court, you want those bounce muscles to take the bulk of the load. But be careful with the aerobic step. Don't try to do too much or injuries become more likely.
    • Coordination: There are a couple of tools we like to train coordination. One is the speed rope. Same caveats as before, don't jump on concrete, and work to do some of the more advanced jump rope exercises like skiers and twisters which emphasize coordination. Coordination will come quickly because you need to engage upper body, lower body, and mid-body all at the same time. We also like the agility ladder for coordination. The agility ladder is a bit boring until you try some of the more advanced patterns and exercises (as shown below), but once you get used to those, you will probably come to like the rhythm. But always remember to engage your arms when running on the agility ladder. Your arms should swing in a rhythm opposite of your feet. When your left knee is up, your right elbow is up. Forcing both the elbow and the knee to rise and fall at the same time is good for coordination. However, like with the jump rope, avoid using an agility ladder on concrete.


    A grass surface is ideal for the agility ladder. An artificial turf surface, or a well-designed running track, is good. A well-sprung hardwood floor is better. Once again, we see cardio-tennis clinics and workshops lay an agility ladder out on a hard court surface, and that is unfortunate because it can cause joint problems. 
    • Flexibility: Yoga is the gold standard for training flexibility. But yoga is a big commitment. It really takes several hours a week for several months to get satisfactory improvement in flexibility. That's in part because we tend to train our muscles to be big in girth, but not long. Yoga trains the muscles to be long. And just as it takes a while for a body building routine to make your muscles bigger around, it takes a while to make your muscles longer. Athletic stretching, especially immediately before (and immediately after) play has somewhat lesser rewards, but the rewards are immediate. To be sure, you should stretch before and after play anyway. It reduces injuries and prepares the muscles and joints for activity. It can be combined with yoga or other flexibility exercises as well. There are different schools of thought on pre-match stretching. Some experts advise dynamic stretching only, and static stretching after the match. The claim is that static stretching, like reaching down and touching your toes, relaxes the muscles and reduces their reactive snap in play. Other experts recommend static stretching before and after play. In our opinion, if you are prone to joint injury, static stretching before play should be mandatory. There is also a different kind of athletic stretching that we've become fond of. It's assisted athletic stretching. Many personal trainers have learned how to stretch their clients. The trainer will use his or her weight and strength to stretch the client. This is good, and if you have a personal trainer you should talk to him or her about it, especially if you have flexibility issues. The luxury option is a specialize stretch therapist. StretchU is probably the most important practitioner of this approach. StretchU is basically like another gym membership, and a fairly expensive one at about $100/month, for which you can get 4 stretches per month. But their physical therapists are highly trained, and if you have serious flexibility issues, it might be worthwhile. By comparison to a typical stretch session with a personal trainer, a StretchU session will be less painful and more taylored to your specific injury or concern, and you will feel more flexibility when you are done. It's like a massage session, except instead of massaging the muscles, they stretch them. 
    • Balance: Yoga is an excellent way to train balance. And yoga delivers results in balance before it returns results in flexibility, largely because you don't need to grow muscles longer to improve balance where you do need longer muscles for flexibility. It's more about muscle memory and learning to activate the fast twitch muscles. But if you can't do yoga, the Bosu Ball is a great tool for training balance, and it's pretty simple. One of the things it does is to get you in touch with your fast twitch muscles right away. If you stand on one foot on the Bosu Ball, after maybe 30 seconds (or maybe 10), your leg will start to shake, and the shaking will increase. That's the fast twitch muscles firing in an effort to keep you in balance on the ball. If you step on the Bosu Ball a couple days later, it won't be as bad because those fast twitch muscles have already built up a memory of what to do. You can also squat on the Bosu Ball.

    Some athletes squat with dumbbells on the Bosu Ball, others simply reach their hands out in front of themselves when they squat down, as in the picture above. If you hold the squat in the lower position, you will feel the fast twitch muscles firing in just a few seconds.
    • Stamina: Running was traditionally the way athletes trained for stamina. For many years, Andre Agassi ran five miles every morning. However, he came to reject that kind of running in his routine. We tend to shy away from traditional distance running to train for stamina because it can be very hard on the knees. Many good endurance athletes have had to have their knees replaced due to the joint damage caused by the constant pounding incurred by running. Bicycles are easier on the body. And hill riding is very serious stamina training. If you can't get on the road, the elliptical or rowing machine at the gym is a reasonable substitute. We particularly like an elliptical trainer called the Long Stride trainer. It allows you to change the length of your stride while exercising. All of these machines emphasize lower body endurance. The muscles in the lower body tend to be the largest muscles. Engaging them triggers greater blood flow and greater movement of oxygen as opposed to engaging only smaller muscles. For similar reasons, the jump rope is also excellent for training endurance. This is one of the reasons boxers depend so much on the jump rope in their training. There are also on-court drills. For most sports there are so called "star drills" where you set out cones in a diamond shape on the court and mark a position in the middle of the diamond. You run to the first cone and return back to the middle, then the second and return, third and return, fourth and return, then repeat. All of this said, many racket sports players will find all of these forms of stamina training too boring. A more interesting approach to training stamina is to go play some squash. Of course, squash players themselves do a great deal of difficult work to build stamina, but the sport requires so much of it that if you are coming from another racket sport, playing a little squash will greatly increase your stamina, and you will see the results very quickly. Badminton may also do the trick. 

    <<More stuff here>>


    This is probably the most fruitful way to train. With your honest self-evaluation in hand, you know which aspects you are strongest in, and which you are weakest. But importantly, each aspect touches on all of the other aspects. The diagram below shows a small sampling of how the different aspects connect.

    So the key to cross-training is to use your strong aspects to up-level your weaker aspects. Say, for example, that you evaluate high in athleticism and low in technique. Ketthong Pisa, Assistant Coach at the San Diego Table Tennis Association, has a powerful way of training through that imbalance. Ketthong is also a Kung Fu blackbelt. His approach, which we call "Kung Fu Racket" uses his strength not so much to move the racket around, but to prevent the racket from going where you don't want it to go. In Kung Fu, you often must take a blow. The opponent's purpose in delivering the blow is to make your body move into a position you don't want it to move to. Your Kung Fu mastery uses your strength and athleticism to absorb the blow without letting the blow move you from your preferred place. 

    Let's go back to our earlier example, where your coach tells you to start your swing "here", then move through the positions "here, here, here, and here", finishing in the position "here". The Kung Fu Racket method uses your strength not so much to move the racket through those positions, but to prevent your body from taking any other positions. In essence, you use your muscles to create a channel in which the only motion possible is along the desired path. And then it is simply a matter of letting your body flow. This indeed requires a lot of strength because the forces on the court--time, motion, the ball, opponent-induced distress--are all trying to pull you away from that preferred swing path, much like an opponent in a Kung Fu fight tries to move you from your preferred path with a blow. We've found that the Kung Fu Racket approach works. So long as you have the strength to start with, it leads to improved technique that reduces any imbalance between athleticism and technique. 

    This is just a small part of one example. Associated with each line in the diagram above are training paths going in each direction. So, just as Kung Fu Racket is a way to use strong athleticism to improve weak technique, there are ways to use strong technique to improve weak athleticism. 

    <<lots more stuff here>>