Water Fast Results. Every Milestone From One Measured 90 Day Fast

Search for water fast results and you get two kinds of page. Studies with no faces in them. Transformation posts with no numbers in them. This page is the third kind. It is every milestone from one doctor supervised 90 day water fast that is still running, pulled into one place, with the morning it happened attached to it.

John is 56, he lives in San Diego and he started at 228.6 pounds on August 1 2026. Same scale every morning, same ketone strip every morning, photographed. His doctor drew labs before day 1, redraws monthly and can stop the whole thing whenever she wants. Nothing on this page is rounded to make a point.

Where the results stand at 40 full days

Every milestone in one table

This site counts mornings. Day 1 is the first morning without breakfast. The morning of day 8 is seven full days and the morning of day 41 is forty. Every weight below is first thing in the morning on the same bathroom scale. The strip column is the Keto Mojo urine strip matched against the chart printed on the bottle, where 0 is a blank pad and 16 is the darkest square it prints.

SWIPE THE TABLE SIDEWAYS FOR ALL COLUMNS →

Look at the one month row. It is the only milestone on this page where the total lost went down instead of up. It is also the most useful row in the table. More on that further down.

The weekly shape, which is the real result

A single day number tells you nothing. One morning in this fast moved 4.4 pounds and another moved zero. Grouped into full weeks the picture stops lying.

SWIPE THE TABLE SIDEWAYS FOR ALL COLUMNS →

Week one is 17.2 pounds and every week after it is between 6 and 8, with one exception that has a cause. That gap is the single most misread thing about water fasting results. Glycogen, the body’s stored carbohydrate, is held with roughly three grams of water for every gram of fuel. When the fuel goes the water goes with it. Nobody burns 17 pounds of fat in seven days. The published reviews of prolonged human fasting describe the same curve, a steep opening drop that settles into a slower steady loss.

What the body scan says

The scale is the loud number. The composition readings are the quiet ones and they came from the same scale every morning.

The muscle line is where fasting pages start lying. Read it slowly. Muscle percent going up does not mean muscle was built. It means fat left faster than anything else. Muscle became a bigger share of a smaller man. Share is not mass. What the number does honestly say is that 40 days without food did not gut his muscle share, which is the fear most people bring to a fast this long. These are consumer scale readings. The harder evidence is blood. That has its own section below.

The BMI line has a date attached to it. John started inside class one obesity at 34.6. On the morning of day 23 his scale read 197.0 pounds and a BMI of 29.9, which crosses the standard obesity line at 30. That is the morning he stopped being classified as obese. He wrote about it in the day 23 note.

The ketone strip result, blank to pinned

The strip is the only daily proof that the body actually switched fuels. John’s arc is worth more than any chart of what should happen.

1. DAYS 1 TO 3Blank

   Three mornings, nothing on the pad. Running on stored carbohydrate. The scale had already given up 8.0 pounds and the strip still showed zero, which tells you what those 8 pounds were.

2. DAY 4First trace

   The pad shows color for the first time at the 72 hour mark. The same morning his energy came back.

3. DAYS 5 TO 7Swinging

   An 8, a dip back to 4, an 8 again. Urine strips move with hydration. Single mornings are noise.

4. DAYS 8 TO 30Pinned at the darkest square

   From the 168 hour mark the pad read 16 every single morning for more than three straight weeks.

5. DAYS 31 TO 35Knocked down

   A small broth and a quarter of an avocado on day 30 dropped the strip to 4 the next morning and then to blank for four mornings in a row.

6. DAYS 36 TO 41The climb back

   A faint 0.5. Then 1.5 on day 37 when the scale moved 2.9 pounds. Then 4 on day 38 and back to 16 on day 39, where it has stayed.

The blood results, both panels published

Weight is the result people search for. Blood is the result the fast is actually for.

• Baseline, Labcorp San Diego, July 31 2026. Drawn fasting the morning before day 1. Exactly one marker came back flagged high, a fasting glucose of 106 against a 70 to 99 reference range. Fasting glucose from 100 to 125 is what the American Diabetes Association calls prediabetes. That single number is the reason this whole project exists. Sodium 140, potassium 4.6, creatinine 1.05, eGFR 83, AST 22, ALT 35, all inside range.
• Month one, at home kit, August 31 2026, day 31. A much wider net, 25 markers. The kit's own scorecard reads 36 percent optimal, 60 percent average and 4 percent out of range. Total cholesterol 111, HDL 44.1, LDL 48.4, triglycerides 92.5, ApoB 61.3, high sensitivity CRP 2.02.
• What the two panels can actually compare. Kidney filtration improved, eGFR 83 to 103.71, with creatinine dropping from 1.05 to 0.804. Alkaline phosphatase came down from 120, near the top of the Labcorp range, to 113. Albumin held steady at 4.9 and 4.95.
• The number still unanswered. The home kit does not test fasting glucose. It measures fructosamine instead, which came back at 231. The head to head rematch on the 106 waits for the next clinic draw, due at the end of September.

Both panels are published in full, typed from the reports with the photos, on the bloodwork page. On day 41 John said on camera that these are probably the best lab results he has had in about 15 years. That is his read on his own blood, not a claim this site is making for him.

The results that never show up on a scale

These are John’s own reports rather than measurements. They are labeled that way on purpose.

• Waist, about 40 inches to a loose 36. Reported on day 34, along with his ring sliding on freely and his size nine and a half shoes fitting with room to spare. The day 34 note has it in his words.
• Shirts, extra large to a tight large. Day 39. He also noticed his upper body coming down faster than his lower half and called the current shape a pear, which is the least flattering and most believable progress report on this whole site.
• The first minute out of bed. Around day 9 he woke with very little of the joint and muscle stiffness a 56 year old body usually reports. One good morning is not a study. It is the thing he says keeps him doing this.
• Sleep. By day 26 he was calling it the best sleep in years.
• Appetite. Gone by day 7 and mostly gone since. On day 41 he drew the distinction that matters, sitting across a restaurant table from a client's steak. He was not hungry. He was tempted. Those are two different problems.

The week the results went backwards

Every honest results page needs the row where the number got worse. This is that row.

On day 30, one third of the way through the 90, John broke 30 days of water only with a small homemade vegetable broth and a quarter of an avocado. He planned it openly on day 28 and reported it the same day he ate it. Here is what one small serving cost.

• His scale went from 190.2 on day 30 up to 192.0 the next morning, the first gain of the entire fast.
• His ketone strip fell from 16 to 4 overnight and then to blank for the next four mornings.
• It took six mornings to get back to 190.2, the weight he was standing on before the broth.
• Week five, the week that contained it, produced 0.9 pounds. Every other full week produced at least 6.4.

Then the dam broke. Day 37 moved 2.9 pounds, his biggest single morning since day 3. His own read was that a week of holding fluid finally let go. The five mornings from day 36 to day 41 gave up 7.7 pounds. His account of the whole thing is in the day 31 note. If that is what a quarter of an avocado does after a month, it is worth respecting on the way out of any fast.

What these results do not prove

• They are one man. One 56 year old starting at 228.6 pounds, under supervision. A lighter person loses less and a different body holds water differently. Nothing here is a sample size.
• They do not settle the glucose question yet. The 106 that started this has not been retested at a clinic. Until it is, the biggest claim this fast could make stays unmade.
• They cannot show you autophagy. The science is real and the 2016 Nobel Prize in Physiology or Medicine went to the work behind it. The hour by hour autophagy timelines all over the internet are not measurements. Nothing in a kitchen confirms one. A scale, a strip and a blood panel are what this site can actually show you.
• They say nothing about what happens after. John regained the weight after a supervised 83 day fast years ago. He says so himself. Results during a fast and results a year later are two different pages. Only one of them has been written.
• They are not a target. The famous 382 day fast in the medical literature is a case report from the 1960s, not a benchmark. This fast is 90 days with a doctor holding a stop switch.

Who should not be chasing these numbers

• Anyone pregnant or breastfeeding.
• Anyone with diabetes or on blood sugar medication.
• Anyone with heart disease, kidney disease or a history of an eating disorder.
• Anyone under 18.
• Anyone on daily medication that assumes food.

Past that list, anything longer than a couple of days stops being a solo project. The risks that actually matter on an extended fast are heart rhythm, electrolytes and careless refeeding. All three are the kind you cannot feel coming. John has a doctor who has known him for years, baseline labs, monthly retests and someone with the authority to end it. That is why he can do what he does. It is not a reason for anyone to do it alone.

Follow the rest of it

The 90 days are not finished. Day 41 is the latest morning on this page and the fast runs to October 29 2026. Every number above updates on the bloodwork page the morning it happens, the daily notes live in the fasting journal and the guides broken out by length sit on the guide hub.