The Novak Djokovic Diet (2026): What He Cut, What He Said, and the Test That Proved Nothing

For years the most talented player in men's tennis kept breaking down in the middle of matches he was winning.

It was not a mystery to anyone watching. He would go up a set, start struggling to breathe, call the trainer, then retire or lose. The 2010 Australian Open quarter-final against Jo-Wilfried Tsonga is the one people remember — he was ill on court and lost in five. Commentators put it down to fitness. Some to nerve. He had one major title, from 2008, and a growing reputation as a man whose body quit on him.

Then he changed what he ate, and 2011 happened: the Australian Open, Wimbledon, the US Open, the world No. 1 ranking. He has spent the decade since crediting a change of diet for a large part of it.

The interesting question is how much of that is verifiable, and one piece of it that is not.

What he actually cut, and when

The change traces to a 2010 meeting with Igor Cetojevic, a Serbian doctor who had watched Djokovic struggle on television. In Serve to Win (2013), subtitled The 14-Day Gluten-Free Plan for Physical and Mental Excellence, Djokovic describes the effect in his own words: "Suddenly there was an X factor, a change in my diet that allowed my body to perform the way it was meant to."

He described the first fourteen days as immediate and physical — sleeping better, waking clearer, feeling lighter on his feet. Not a slow build. A change he noticed inside a fortnight.

He also said it publicly at the time. At a press conference in Belgrade: "He's done a great job in changing my diet after we established I am allergic to some food ingredients like gluten. It means I can't eat stuff like pizza, pasta and bread."

And he kept doing it. The diet outlasted the season that produced it, and the book came two years later.

So the habit is well documented, in his voice, across twelve years. Gluten out. Dairy out. Tomatoes reduced. That part is not in dispute.

The test that started it, and why we are not repeating it as fact

Here is where most articles about this stop being useful.

The way Cetojevic demonstrated the problem, as Djokovic describes it, was a muscle test. Hold a slice of bread against your stomach with one hand. Hold the other arm straight out. The doctor presses down on it and compares how strong it feels with the bread and without. Djokovic's arm was weaker holding the bread. His own summary: "there was a noticeable difference."

That procedure is applied kinesiology, and it is not a diagnostic test for food intolerance. A review of the literature concluded there was insufficient evidence for diagnostic accuracy or for the validity of the muscle response for any condition (Hall and colleagues, 2008; 22 studies, all of poor methodological quality). Where reliability has been measured, tests using non-musculoskeletal challenges — exactly what "does this bread weaken you" is — come out at essentially no reliability.

Djokovic has also never been reported as a diagnosed coeliac. He is gluten-free by choice and by how he says he feels, not by biopsy or serology.

We are blunt about this for a reason. Read the arm test as a way to find out what your body doesn't like and you will get a wrong answer with total confidence, then cut foods you had no reason to cut. The habit is worth understanding. The method he was sold on is not.

What is left standing is unglamorous and real: he removed two large categories of food, ate that way for a long stretch, paid close attention, and kept what seemed to work. An elimination trial run by an athlete on himself, with all the limits of one person watching himself while expecting to feel better.

The habit, precisely

1. Two categories out, not a hundred rules. Gluten and dairy, gone; tomatoes a lesser sensitivity. A short list, which is why it survived fifteen years of hotel rooms.
2. A fixed two-week test, then a verdict. The book is built around fourteen days. Not "eat better" — a defined window with an endpoint where you decide.
3. The same food most days. Vegetables, beans, white meat, fish, rice, non-gluten grains. Boring on purpose. Elite diets are usually a small rotation repeated, not variety.
4. Attention paid every day. What he keeps describing in interviews is not a nutrient. It is noticing — how he slept, how he felt on court, what preceded a flat day.

What the science says

Non-coeliac gluten sensitivity is real as a reported experience, and hard to pin down as a mechanism. People report symptoms from gluten-containing food without coeliac disease or wheat allergy. Double-blind studies keep finding that many of them do not react to isolated gluten when they don't know it's there — which points at other components of wheat, or at expectation, or both. Djokovic's experience is compatible with a real improvement and with gluten not being its cause.

Cutting gluten is not, by itself, a performance intervention. There is no good evidence it helps athletes who don't react to it. What it does in practice is remove a lot of bread, pastry, beer and packaged food, and replace them with things you cooked.

Cooking at home tracks with better diet quality. A study of 9,569 US adults found people who cooked dinner at home more often ate fewer calories, less sugar and less fat (Wolfson & Bleich, Public Health Nutrition, 2015). Cross-sectional, so it cannot say cooking caused it. But a rule that makes most restaurant and convenience food unavailable pushes you into your own kitchen, whatever the rule was nominally about — the same pattern we trace across everyone who cooks their own food on the record.

Writing down what you eat is the best-evidenced piece of any of this. In a trial of 1,685 adults, the number of daily food records people kept was the strongest single predictor of how much they changed (Hollis et al., American Journal of Preventive Medicine, 2008). The caveat matters: everyone was also in a supported group programme, so the records were not working alone. And a meta-analysis of 122 studies found interventions prompting self-monitoring outperformed those that did not (Michie et al., Health Psychology, 2009). Whatever you conclude about gluten, the noticing does measurable work.

A new way of eating takes longer to feel normal than people expect. In the standard habit study, automaticity took a median of 66 days among the 39 participants whose curves the model fitted (of 96 enrolled), range 18 to 254 (Lally et al., 2010; self-reported). Fourteen days is enough to feel a difference. Not enough for the new way to stop requiring thought.

What this story cannot prove. Djokovic changed his diet and then had a historic season. Two facts in sequence, not a chain. In the same period his serve was rebuilt and Serbia had just won the Davis Cup — which he has himself described as the confidence turning point. Either could carry as much weight as the food. He also has a full-time chef and generational talent. Thousands go gluten-free and become the best in the world at nothing. The claim that survives is small: he ran a defined dietary experiment on himself, watched carefully, felt better, and kept it.

Why the short list is the part that works

Most diets fail because they are a hundred decisions a day. Djokovic's is two lines long, and two lines you can hold in your head at a wedding or an airport.

There is a second thing hiding in it. A rule that removes bread and pasta and cheese removes most of the food that gets eaten without anyone deciding to — the roll on the table, the leftover sandwich, the pizza somebody ordered. You are not fighting those with willpower. They stopped being options.

How to run it yourself, starting tonight

• Pick one category, not five. He cut gluten first, then dairy. Remove one thing you eat most days. Cut six at once and if you feel better you will never know why.
• Set a fourteen-day window with a date on it. The book's structure is a bounded test. Write the end date down. A test with no end date is a mood.
• Write two lines every night. What you ate, and how the day felt. Ten seconds. That record is the part with 1,685 people and 122 studies behind it, not the food rule.
• Decide on day fifteen, then put it back. Reintroduce the food deliberately for a few days and see if the feeling reverses. That step separates an experiment from a belief.
• Get a real test if the answer matters. If you suspect coeliac disease, get tested before you cut gluten — the blood test and biopsy need gluten in your diet to work, so going gluten-free first can make a real diagnosis impossible for months.

The mistakes that kill this habit

Copying the arm test. The most repeated detail in every article about him, and the least reliable thing in the story. Muscle testing against food is not a diagnostic (Hall et al., 2008; 22 poor-quality studies).

Cutting everything at once. Gluten, dairy, sugar, caffeine, all on a Monday. You will feel strange for a week, learn nothing, and be back to normal by Sunday.

Going gluten-free before getting tested. If coeliac disease is a real possibility for you, this ordering error costs you a real answer. Test first.

Swapping in gluten-free packaged food and calling it done. A gluten-free cookie is a cookie. If the change works, it is probably because more of your food came out of a pan.

Judging it on four days. Fourteen was the number in the book, and even that only tells you how you feel, not whether it lasts (Lally et al., 2010; the 66-day median is over 39 of 96 enrolled).

Reading a diet as a treatment. Persistent gut symptoms, weight loss you did not intend, or fatigue food does not touch are a conversation with a doctor.

Where HealthBrew fits

The transferable part of this is not the food list. It is that he ran a bounded experiment on himself and paid enough attention to notice the result.

HealthBrew's nightly close is that attention as a sixty-second habit. How the day went, what you ate that was different. After a few weeks you can see which days came before your good ones — the thing you cannot see from inside a single Tuesday. Your first 30 daily closes are on us, no credit card.

For the other half of his routine, we wrote separately about the fifteen minutes of daily meditation he describes in the same book.

One last honest note. We wrote about Djokovic because he won twenty-four majors, and picking people because they won is a biased way to pick anyone. Plenty of players changed their diet in 2010 and are in no record book.

Educational content, not medical advice. Quotes are Djokovic's own, from Serve to Win (2013) and his Serbia Open media day in Belgrade, April 2011. The applied-kinesiology arm test described in the book is not a validated diagnostic method and is reported here as part of the story, not as guidance. If you suspect coeliac disease, get tested before removing gluten — testing requires gluten in your diet. Persistent digestive symptoms deserve a doctor.