Something changed between week two and week six of the cut. Your arms sit differently inside a sleeve. Your chest looks flatter under the same lighting that used to catch the definition. When you flex, the contour that was there a month ago has softened. The observation is real, measurable, visible — and why muscles look flat when dieting is a question that tends to arrive right around the moment the mirror starts disagreeing with the effort.
Most people read this as muscle loss. Calories dropped, the deficit started consuming tissue, and the months of building are being reversed. The fear writes the diagnosis before any data arrives. It is half right. Something did leave the muscle cell. Not all of it was muscle.
Why Muscles Look Flat When You're Dieting
Inside every muscle fiber, stored carbohydrate sits bound to water. That combination — fuel plus the fluid locked to it — accounts for a visible portion of what muscles feel like when they are full. A caloric deficit reduces the carbohydrate available to replenish those stores. Glycogen drops. The water it held leaves too. Previous research has shown that this loss can shrink muscle volume for a while without affecting the contractile tissue. What leaves is volume, not structure.
When you diet, reduced carbohydrate intake depletes glycogen stores inside your muscle cells. Glycogen is stored with water, and previous research shows their loss can transiently reduce muscle volume without affecting contractile tissue. In one trial, a week of carbohydrate reintroduction after ten weeks on a ketogenic diet raised fat-free mass by 4.8%, a gain the meta-analysis authors call likely attributable to glycogen repletion.
— Henselmans et al. 2026 · Sports Medicine · 11 RCTs
It can reverse when the glycogen comes back. After ten weeks on a ketogenic diet, participants who reintroduced carbohydrates for a single final week saw their fat-free mass jump by 4.8%. The control group, eating carbs throughout, showed no change in that final week. The meta-analysis authors call that jump likely attributable to glycogen repletion: a scan of fat-free mass counts glycogen and water along with muscle.
The muscles did not shrink. They deflated.
Training compounds the effect from both directions. A single resistance session drains roughly 39% of the glycogen stored in the working muscle. In elite athletes eating 8 grams of carbohydrate per kilogram a day or more, with only moderate muscle damage, that tank refills within 24 hours. On a cut — when carb intake is already reduced — replenishment runs behind. Each session depletes. The diet slows the refill. Muscles stay flatter between workouts because the fuel inside them may not fully return before the next training day.
How much visible mass actually disappears depends on how far carbohydrate drops. Under tightly controlled conditions, near-zero carb intake stripped 1.61 kilograms of fat-free mass in fourteen days. Not fat. Not contractile tissue. Glycogen and the water that was holding onto it.
None of this is static. Glycogen rises after a carb-heavy meal and falls after you train. The flatness following a leg day and a low-carb dinner is more pronounced than what you see after a rest day and a plate of rice. The mirror reports a moving signal, not a fixed verdict.
The growth underneath is a separate question. Across eleven randomized trials, carbohydrate intake made no significant difference to how much muscle people built during resistance training, though the authors rate their certainty low, and those trials tracked growth across full training programs, not the day-to-day shifts a mirror shows. The visual loss and the growth trajectory run on independent tracks. One is a fuel gauge. The other, a construction project.
Underneath the flat surface, the glycogen loss itself changes nothing structural. The role carbohydrate plays beyond visual fullness extends beyond what the mirror can measure. The part the mirror was reporting on is simpler. When the deficit ends and carb intake rises, glycogen returns, water follows, and the volume fills back in — measured in days, not months. Every mirror check between now and then is reading the fuel gauge, not the structural plan.