Can you spot reduce fat? (Do ab exercises burn belly fat?)

The evidence-led answer

NST verdict

No, you cannot train a body part to remove the fat sitting over it. Two randomised trials tested this directly. In one, people trained a single arm for 12 weeks; scans showed fat came off the whole body, not preferentially from the trained arm. In another, six weeks of daily abdominal exercises on a steady diet produced no change in belly fat, waist size or body fat percentage, only better ab endurance. Fat is released from stores all over the body when you are in an energy deficit, and burned wherever energy is needed; there is no pathway that routes it from the fat cells to the muscle just beneath them. Where you lean out first is mostly down to genetics and sex. Visible abs are a whole body fat outcome, not a crunch count outcome. Train the muscle underneath for strength and shape, just not as a fat loss trick for the skin on top.

Does training a body part burn the fat over it?

No, and this has been tested directly, twice. One arm, 12 weeks: 104 people did a supervised resistance training programme with their non dominant arm only. Measured by MRI, subcutaneous fat decreased across the whole body, not preferentially in the trained arm. A skinfold reading suggested a local effect in men, but the authors put that down to measurement noise, since a single skinfold at one spot is far less reliable than an MRI along the whole arm. Abs, 6 weeks: a group did seven abdominal exercises, five days a week, on a diet with calories held steady. There was no change in body weight, body fat percentage, abdominal fat, waist circumference or abdominal skinfolds versus a control group. The only thing that improved was abdominal muscular endurance.

Why not?

Because fat is mobilised from all over. When you are in an energy deficit, your body releases fat from stores everywhere, through the bloodstream, and burns it wherever energy is needed, which is mostly not the muscle you just worked. There is no mechanism that connects "this muscle is contracting" to "burn the fat cell directly above it". Working a muscle increases blood flow and gives a short lived pump, but that is not fat loss.

So why is one area always "the last to go"?

Genetics and sex, mostly. People have individual patterns of where fat is stored and in what order it comes off, abdomen and lower back for many men, hips and thighs for many women. Training does not override that pattern. An area feels "stubborn" because it is simply later in your personal sequence, not because you are training it wrong.

How do you actually get visible abs (or lean out an area)?

By lowering your overall body fat, through a sustained energy deficit; the body recomposition article covers how. The abdominal muscles are already there. They show when the fat layer over them is thin enough. That threshold is different for everyone and there is no single body fat percentage that guarantees it. Crunch volume does not change the fat layer. Only the deficit does.

Should you still do ab / arm / leg work?

Yes, for the muscle itself. Direct work builds the strength, size and shape of the muscle underneath, which affects how the area looks once body fat is low, and supports posture and performance. Just do it for those reasons, not as a way to burn the fat on top.

What this means in practice

If your goal is to lean out a specific area, the lever is overall fat loss: set up a modest calorie deficit, keep protein high, and keep training. Do direct work for the muscles you care about because it shapes the muscle, and accept that the order your body sheds fat is not something you control. Skip "targeted fat loss" belts, heated wraps, vibration plates and apps; they have marketing, not a mechanism. If a particular body area has become a fixation that is affecting how you feel day to day, that is worth raising with a qualified professional.

Limitations and uncertainty

Both trials are small to moderate and short, 6 and 12 weeks. They are direct tests, they agree with each other, and they agree with the physiology, but a longer or larger trial would strengthen the picture. The "genetics and sex determine the order" point is well established in fat distribution research but is presented here as synthesis, not from these two trials. This is about subcutaneous fat and appearance, not visceral fat or metabolic health specifically.

What would change our conclusion

A carefully run trial showing a clear, MRI or DXA measured local fat loss over a trained muscle that exceeds the loss elsewhere. Or evidence of a physiological pathway linking local muscle contraction to local fat oxidation at a meaningful rate.

How NST assessed this

NST ran no studies. This article is built from published research: a 12 week randomised trial of single arm resistance training with MRI measured subcutaneous fat, and a 6 week randomised trial of abdominal exercise with a calorie matched diet. Bibliographic details were checked against PubMed in September 2026. Each claim is tied to a named source, and the underlying evidence records still need a human editor to read the full texts, assign a plain language certainty, run the tone check, and confirm nothing newer changes the picture.