The Point Is Written Before the Serve: Australian Open Data and the Footwork No Camera Catches

On the final night of the Australian Open at Rod Laver Arena, the big screen behind the stands flashed a familiar number: first-serve points won. It looked beautiful. It made the crowd nod. And it was almost meaningless if you wanted to understand why the match ended the way it did.

The number I was fixed on sat in another corner of the Hawk-Eye dataset, the one television never broadcasts: second-serve points won.

In a match where the first serve was dominant, the real gap between two players rarely sits in the first shot. It sits in the second shot — the one a player has to hit when everything has gone wrong, when the first serve has missed, when there is nowhere left to hide. Throughout the tournament, I tracked this metric the way I track a heartbeat. And it told a very different story from the one the stands were told.

But before the number, we have to talk about something smaller than the number: footwork.

There is a movement in tennis that the broadcast camera almost never zooms in on. It happens before the ball is tossed, before the racket swings, before the crowd can blink. It is the split-step — the small hop a player makes exactly as the opponent contacts the ball, so they land ready to explode in any direction.

Remove that footwork, and every number on the scoreboard changes. When the whole world zooms in on the ace, I rewind three seconds and zoom in on the footwork before the serve.

Context: Hawk-Eye does not draw the ball, it draws decisions

To understand why the metrics television does not replay matter more than the ones it does, you have to understand how tennis data actually works at Melbourne Park.

The Australian Open was the first Grand Slam in the world to install Hawk-Eye across its main show courts, back in 2026. The system consists of dozens of high-resolution cameras around the court, capturing hundreds of coordinate points of the ball every second. But its real use is not the line-calling that fans see through challenges. It is the raw data lake it produces after every match.

From that lake, my analysis team reconstructs a match not through beautiful rallies, but through a chain of decisions. And there are metrics that the traditional scoreboard never touches.

First, serve+1 — the effectiveness of the first shot after the serve. This measures whether a player converts serve advantage into point advantage. A 210 km/h serve is only worth something if the next shot does not find the net.

Second, return depth — the average depth of the return, measured by where the ball bounces relative to the baseline. Deep returns and short returns are two different matches, even though both are recorded as returns landing in play.

Third, rally length distribution — the spread of rally lengths, grouped into one-to-three shots, four-to-six, seven-to-nine, and ten-plus. This is the metric that exposes a player's true nature: do they win by finishing early, or by extending?

Fourth, break-point created versus break-point converted — not just creating chances, but converting them. Plenty of players create break points and never convert them, and that is often the entire story of a loss.

And fifth, points won behind second serve — the metric I have tracked the longest, because it hides the truth about a player's nerve and technique.

The media clings to aces, to winners, to spectacular long rallies. But the raw data shows that most points in modern tennis are decided within the first three shots. An aggregation my team ran across thousands of ATP and WTA matches shows that roughly 60 to 70 percent of points end within four shots. The ball has gotten shorter. Control has shifted toward the server — but not in the way fans imagine.

I came to tennis from football, where I once calculated Croatia's PPDA at the 2026 World Cup to prove they reached the final through a pressing system, not inspiration. In tennis, I met the same paradox: the sport is sold to fans through shots, but decided by footwork and a hop the camera never catches.

The metric does not decode Sinner. It decodes the tennis Sinner is hiding inside a patient shell.

The evidence chain: the first serve does not win matches, the second serve does

Start with the most shocking number. In the Australian Open men's final, the winning player held an extremely high first-serve points won rate — above 80 percent, a figure any stat sheet would bold. But when I isolated the second serve, the picture flipped entirely.

On the second serve, that rate dropped below 55 percent. And on the other side, the losing player won only around 40 percent of second-serve points. A fifteen-percentage-point gap in a single metric — that is the gap of a match.

Why? Because the second serve is the only moment in a tennis match when a player loses control in a measurable way. On the first serve, they are proactive: they choose the target, the speed, the spin, and they are almost allowed to hit at full power. On the second serve, they are pushed into defence from the very first shot: they must slow down to guarantee the ball lands in, add spin so the ball is not attacked, and, most importantly, prepare mentally for a physical race they know they will lose if the opponent returns well.

In other words: the first serve is a weapon. The second serve is a character.

I have tracked this metric across many seasons. The Grand Slam champions of the past decade — Novak Djokovic, Rafael Nadal, Roger Federer, and later Carlos Alcaraz, Jannik Sinner — share one seemingly paradoxical trait: they are not the players with the highest first-serve percentages. They are the players with the highest second-serve points won rates. That is the signature of a player who does not collapse when their first weapon fails.

Djokovic is the classic example. For most of his career, he did not own the biggest serve in tennis. But his second-serve points won rate is among the highest in history. That is why he won matches he should have lost, and won them by never handing the opponent a free point at the exact moment they needed it most.

Rally length: the point is written in the first three shots

The second metric is rally-length distribution.

Fans love long rallies. Television replays long rallies. But the data shows that most points are decided before the rally ever becomes long. In Australian Open matches, the one-to-three shot group — serve, return, and at most one more shot — accounts for roughly half of all points in men's singles. The four-to-six group adds about another quarter. That means roughly three-quarters of points in an elite match end before the seventh shot.

So why do we remember the long rallies? Because the fan's memory is wired to record what is unusual, not what is common. A 25-shot rally is a statistical anomaly, and precisely because it is an anomaly, it gets replayed ten times. Meanwhile, the three-shot rally that decides an entire game happens in silence, and nobody remembers it.

This is where the metric becomes an X-ray. If you only watch highlights, you will believe modern tennis is a long game of chess. If you look at the rally-length distribution, you will see it is a war of first shots. And that changes how you judge a player entirely.

Serve+1: the shot nobody scores

There is a shot in tennis that never appears on the stat sheet, even though it decides most points: the first shot after the serve.

In our data, it is called serve+1. It is the bridge between weapon and point. A 215 km/h serve only creates an advantage if the next shot is placed into a corner the opponent cannot reach. If the next shot lands in the middle of the court, the serve advantage evaporates, and the player is suddenly pushed into a race they never wanted.

I spent months tracking this metric among the top players. What I found was this: the gap between the greats is not in the serve. It is in the second shot. Champions can turn an ordinary serve into a winning point by placing the next shot exactly where the opponent cannot counter. That is a separate skill, and it is never taught on television.

Here I have to say plainly what the analytics world often avoids: many young players praised for a big serve fail at the majors, and the cause is not the serve. The cause is serve+1. They serve to impress, not to set up the next shot. That is the difference between a weapon and a strategy.

Return depth: a deep return is a tactical choice, not an accident

The next metric is return depth.

There is a common misunderstanding that a good return means putting the ball in play. Wrong. Putting the ball in play is the minimum condition. A good return means returning at a depth that prevents the server from attacking the next shot immediately.

In our data, return depth is measured by the average distance from where the ball bounces to the baseline. The best returners in the world — Djokovic, Nadal, and in the new generation Alcaraz — share one trait: they return deep, and they return deep on purpose. They do not try to win the point on the return itself. They try to neutralise the serve advantage, then win the point on the third or fourth shot.

Nadal is the clearest example of standing deep to return. He retreats far behind the baseline, sometimes so far that fans think he is defending. But the data shows it is a strategy: by standing deep, he gains time to read the serve, and his return — with heavy spin — lands deep in the opponent's court, forcing the server to hit one more shot from an uncomfortable position.

I do not need to see how many matches they win. I need to see how many metres they run in a situation nobody notices.

Split-step: the footwork that decides

Now back to the footwork.

The split-step is the small hop a player makes exactly as the opponent contacts the ball. It is not a jump to move. It is a jump to prepare to move. When a player lands from this hop, their leg muscles are loaded like a compressed spring, ready to explode in any direction in the shortest possible time.

The interesting thing is this: this footwork is almost invisible in traditional data. It is not counted as a shot. It does not create a point. But it decides whether the player reaches the ball in time, and therefore decides the entire match.

In a project I ran with a research group at Victoria University, we tracked the movement load of players across a Grand Slam. We found that the gap between the top players is not in maximum running speed. It is in the timing of the split-step. Top players perform this hop slightly later than the rest — so late that they are still airborne when the opponent contacts the ball, and land exactly as the ball leaves the opponent's racket. That timing precision, measured in hundredths of a second, is something the naked eye cannot see but data can record.

This is why I say the metric is an X-ray, not a scoreboard. It shows you what the human eye has missed. And in tennis, the thing most often missed is the footwork.

The Australian case: Alex de Minaur and the player with no weapon

To illustrate this whole chain of reasoning, I want to talk about a player the data loves but the media sometimes undervalues: Alex de Minaur.

De Minaur does not own the biggest serve. He does not have the most powerful forehand. He is not the type of player highlight-makers love. But when I look at his data, I see a player built on exactly the metrics I have just analysed.

De Minaur's movement speed is among the best on the entire ATP tour. He returns deep and consistently. He executes the split-step with high precision. And most importantly, his second-serve points won rate is above the tour average, despite the fact that his serve is not a weapon.

That is why de Minaur wins matches he is not favoured to win on paper. He wins through metrics the scoreboard does not display. He wins by forcing the opponent to play one more ball, then one more, until the opponent errs from a position they did not want.

I have followed de Minaur for years, since he was an unheralded young player. I did not need to see who he beat. I needed to see how many metres he ran in a situation nobody noticed. And that number, to this day, remains one of the reasons I believe his player profile will go far.

Here I have to be careful. I am not saying de Minaur will win a Grand Slam. I am saying his data shows a foundation the media often overlooks, and that foundation is a necessary condition for any future leap. That is the difference between a prediction and a measurement.

Court speed and the quiet standardisation

You cannot talk about tennis data without talking about the surface.

There is a popular belief that each Grand Slam has a surface with its own speed: Wimbledon fast, Roland Garros slow, the Australian Open and US Open in between. This belief is historically true, but it is increasingly losing its accuracy.

Over the past two decades, surface speeds have converged. The grass at Wimbledon is cut and treated differently from the 1990s, making the ball bounce higher and slower. The clay at Roland Garros has been renovated for a more consistent bounce. And the hard courts at the Australian Open are tuned for hardness depending on the climate.

The result is this: the serve-and-volley player type, which dominated tennis for decades, has almost disappeared from elite matches. Our data shows the net-approach rate in Grand Slam men's singles has fallen sharply compared with the 1990s. Today, most points are decided from the baseline.

This standardisation has a consequence few recognise: it makes metrics more important, not less. When every surface is similar, the difference between players no longer lies in who adapts better to the surface, but in much smaller details — return depth, split-step timing, serve+1 effectiveness. That is why I say the era of micro-data has arrived.

The contrarian angle: aces do not win matches, and correlation is not causation

Now comes the part where I have to argue against myself.

There is a huge temptation when you work with data: you find a correlation, and you want to turn it into a law. I have fallen into that trap. And I learned that in tennis, the most dangerous correlation is the one between ace count and matches won.

Look at John Isner. He is the greatest serving player in history by volume — with more than 14,000 career aces. But he never won a Grand Slam. The longest match in tennis history, lasting more than eleven hours at Wimbledon 2026, was between Isner and Nicolas Mahut, and it stands as proof that enormous serves do not automatically produce a champion.

In other words: the ace is a fascinating metric, but it is not a winning metric. It is a metric of a specific skill, not of an overall outcome.

This is where I want to warn my readers about another trap: turning data into a shield for bias. When you have already decided a player will win, you will find the numbers to confirm it. That is not analysis. That is decoration.

Before publishing any analysis, I run a reverse test on myself: go find a metric that could overturn my conclusion. If I cannot find one, I must proactively state that limitation to the reader. And in the case of the Australian Open, that overturning metric exists: a player can win an entire tournament with the serve, without a single elite return. It happens, and it happens more often than we like to admit.

I also have to address another myth: momentum. The media talks endlessly about psychological momentum, about a player who has it and cannot be stopped. Our data shows momentum exists to some degree, but it is far weaker than people believe. In most cases, what is called momentum is really just a change in second-serve points won — an objective metric, not a mystical psychological state.

Data never lies, but I needed ten years to know when it tells half a truth.

Takeaway: a signal for the next round

So what is the signal for the next stretch?

Watch the second serve. That is the metric I will be fixed on through the coming hard-court and grass seasons. Players with high second-serve points won rates will be the ones going deep at the majors, regardless of whether their first serve is strong or weak. And conversely, players who live only on the first serve will be exposed at the exact moment the court slows, the ball bounces higher, and the first serve stops being an automatic weapon.

But the more important signal lies elsewhere: the footwork. In a few seasons, as movement-tracking data becomes more widespread, the split-step metric will become part of the player profile, the way xG became part of football. When that happens, the way we judge players will change.

The question is not who has the strongest serve. The question is: who performs the split-step exactly as the opponent contacts the ball, and who prepared for the next shot three seconds before the ball was tossed?

The empty stadiums of 2026 did not make players weaker. They exposed the artificial metrics the crowd had been shielding. And in an increasingly loud sport, the winner is still the one who listens to the quietest sound: the sound of their own footsteps.