Fire a maximum-effort sprint out of the blocks and the ATP already sitting inside your leg muscles is gone in roughly two seconds. Not depleted in the sense of low. Gone in the sense of used, hydrolysed, spent — the phosphate bonds snapped to release the energy that drove the first two or three strides. And yet the sprint keeps going. For another six, eight, sometimes ten seconds, the legs continue firing at nearly full power. What bridges the gap is a second molecule stashed inside the same muscle fibres at concentrations somewhere in the vicinity of 120 millimoles per kilogram of dry muscle: creatine phosphate, the cell’s emergency reload system.
The number sounds abstract. The consequence is not. It is the reason a 100-metre sprinter can hold peak velocity from about metre 30 to metre 60 without their legs simply stopping. It is the reason a heavy deadlift is possible at all. And it is the reason a supplement first isolated from meat broth in 1832 has become, by a wide margin, the most-studied ergogenic aid in sports science.
The two-second cliff
Adenosine triphosphate is the molecule every cell in the body spends to do work. Muscle contraction, nerve firing, the pumping of ions across a membrane — all of it runs on ATP hydrolysing to ADP and releasing a burst of chemical energy. The catch is that muscle cells do not store much of it. Resting concentrations sit at roughly 5 to 8 millimoles per kilogram of wet muscle, which is enough fuel for a couple of seconds of maximal contraction and no more.
Work by exercise physiologists tracking energy pathways during repeated sprints has mapped this out in detail. A Nature-published analysis of bioenergetic pathway contributions across single and repeated sprints shows how quickly the stored ATP pool falls and how sharply the phosphocreatine system takes over within the first few seconds of effort.
If ATP were the only fuel, a sprint would end before the athlete cleared the third stride.
What Chevreul boiled out of meat
In 1832, the French chemist Michel Eugène Chevreul reduced skeletal muscle broth down to a residue and isolated a compound nobody had described before. He named it after the Greek kreas — flesh — because flesh was where he found it. Almost two centuries later, that etymology still tells you where the molecule lives.
Around 95 per cent of the body’s creatine is packed into skeletal muscle, according to clinical summaries of creatine biochemistry. Most of it sits in a phosphorylated form — creatine phosphate, sometimes called phosphocreatine — bound to a high-energy phosphate group and waiting.
The body makes about a gram of it per day on its own. The liver, kidneys and pancreas assemble it quietly from three amino acids: glycine, arginine and methionine. The rest of the daily supply, historically, came from the dinner plate. Roughly two to five grams sit in a pound of raw beef or salmon. Cook it, chew it, absorb it, and it ends up in the same muscle depot.
The reload mechanism
Here is where the 120 millimoles per kilogram matters. When ATP hands off its terminal phosphate to power a contraction, it becomes ADP. Left alone, ADP is a dead end — the cell cannot use it for work. But sitting next to it in the sarcoplasm is creatine phosphate, holding its own high-energy phosphate bond. The enzyme creatine kinase catalyses the transfer in essentially one step: phosphocreatine plus ADP becomes creatine plus ATP.
The reaction is fast. Faster than glycolysis. Far faster than mitochondrial oxidative phosphorylation, which needs seconds to spin up and requires oxygen the sprinting muscle does not yet have in sufficient supply. Creatine phosphate is the only system in the cell that can rebuild ATP at the rate a maximal contraction burns it.
Think of it less as a fuel tank and more as a bank of pre-charged batteries wired in parallel with the primary supply. When ATP drops, the batteries dump their charge into the depleted pool. No oxygen required. No lactate produced. No wait.
The reservoir is finite. At roughly 120 millimoles per kilogram of dry muscle — the figure cited across the sports-science literature and reproduced in reviews of creatine’s role in cellular energetics — the phosphocreatine store can sustain maximal output for something on the order of eight to ten seconds. After that, glycolysis has to carry the load, and the burn in the quads begins to register.
Why the sprint has a shape
Watch a 100-metre race in slow motion. Acceleration builds through about the first 30 metres. Peak velocity holds from roughly metre 30 to metre 60. And then, on almost every runner in the field, velocity begins to decay. Nobody accelerates through the tape. The winner is often the person who decelerates least.
That shape is the phosphocreatine curve made visible. The first two seconds are stored ATP. The next six to eight are creatine phosphate handing off phosphate groups as fast as creatine kinase can turn the reaction over. Somewhere around the eight-to-ten-second mark, the reservoir runs low enough that glycolysis has to take a larger share. Glycolysis is slower per unit time and produces hydrogen ions as a by-product. The muscle acidifies. Force output drops.
The nervous system knows this. The brain appears wired to calculate the length of an all-out effort in advance, pacing motor output against the fuel it expects to have available. The two-second ATP cliff and the ten-second phosphocreatine limit are among the constraints that calculation is built around.
Why the supplement works — for some people more than others
Creatine monohydrate has been through several hundred controlled trials since it entered the sports-science literature in the early 1990s. The consensus, summarised in healthline’s review of the exercise-performance evidence, is that supplementation raises intramuscular creatine stores by somewhere between 10 and 40 per cent, and that this translates into measurable gains of 1 to 15 per cent in high-intensity, short-duration output.
The size of the jump depends heavily on where a person started. Muscle has a ceiling — the 120-mmol/kg figure is close to a saturation point, not a resting baseline. Someone whose diet already delivers several grams of creatine a day from beef, chicken, pork or fish sits closer to that ceiling and has less room to move. Someone who eats no meat at all can be measurably below it. Give both people the same 5-gram scoop and the second person’s muscles have more empty space to fill.
The performance data reflect this. Studies cited in the healthline review report a 3.7 per cent improvement in cycling power after a four-day loading protocol, faster 40-metre sprint times, and improved 5- and 15-metre acceleration in soccer players. Strength trials show gains of around 15 per cent in one-rep max bicep curl and significant increases in squat and bench press over six weeks — the kind of numbers that would otherwise take months of training alone.
Loading and the saturation ceiling
The classic loading protocol — 20 to 25 grams per day split into five-gram doses across five to seven days, then a maintenance dose of 3 to 5 grams — exists specifically to push muscle stores toward that saturation ceiling as quickly as possible. Lower daily doses reach the same endpoint. They just take three or four weeks instead of one.
A recent analysis reported by News-Medical on the interaction between creatine supplementation and resistance training reinforces a point that is easy to miss: the supplement raises the size of the phosphocreatine reservoir, but the reservoir only matters if the muscle is being asked to draw from it. Without the training stimulus — the sprints, the heavy sets, the plyometric work — the extra creatine sits there unused.
A separate line of research, summarised by Medscape’s review of creatine’s readiness for clinical applications, has begun probing whether the same reload mechanism matters in tissues other than muscle. Neurons also run on ATP. Cognitive tasks that impose acute energetic demand — working memory under sleep deprivation, for instance — show small but measurable improvements in supplemented subjects. The mechanism appears to be the same molecule doing the same job in a different cell type.
What the numbers add up to
The arithmetic of a maximal effort is unforgiving. Resting ATP concentrations in the range of 5 to 8 mmol per kilogram of wet muscle. Phosphocreatine at roughly 15 to 20 times that concentration in the same tissue. A reload reaction that runs in a single enzymatic step. A saturation ceiling that dietary intake can nudge but not exceed.
Every hundred-metre final, every last set of a squat workout, every dash across a road to make a green light rides on this arrangement. The first two seconds spend what is already in the till. The next eight are the creatine kinase reaction firing over and over, faster than the eye can register, refilling the ATP pool from a reservoir the body has been quietly topping up since the last meal.
Chevreul, boiling down his meat broth in a Paris laboratory in 1832, had no idea what he had isolated. He named it after flesh because flesh was where it lived. Two centuries later, that origin still holds. The molecule sits in the muscle. Waiting. And when the gun goes off, it is already moving before the runner has taken a second stride.