Cardio, if the only measure is fat mass: in the best-known head-to-head trial, adults who did cardio for 8 months lost clearly more fat than adults who lifted weights, while the lifters gained lean mass instead1. Larger reviews agree that both work, that cardio moves the scale more, and that doing both covers fat and lean mass together2.
So the honest answer is "cardio for fat, weights for lean mass". The details are below.
What was studied
The trial is called STRRIDE AT/RT. Researchers at Duke and East Carolina universities took sedentary adults with overweight or obesity (BMI 25 to 35, average age about 50) and assigned them by chance to one of three 8-month programmes1. Assigning by chance is what makes it a randomised trial: the groups differ only in the programme.
- Weights (resistance training): lifting against a load that goes up as you get stronger. Three days a week, three sets of 8 to 12 reps on eight machines, about 180 minutes a week.
- Cardio (aerobic training): steady work that keeps you breathing hard, on a treadmill, elliptical or bike at 65 to 80% of peak fitness. The dose matched the energy of jogging about 12 miles a week, about 133 minutes a week.
- Both: the two programmes together, almost double the time.
Nobody was put on a diet. Reported food intake did not change in any group1. So this is exercise alone.
| What | Number |
|---|---|
| Adults analysed | 119 (68 women, 51 men) |
| Length | 8 months |
| Weights | about 180 min a week |
| Cardio | about 133 min a week |
| Diet | not changed |
| Body composition | air-displacement or DXA scan, before and after |
Fat mass is the fat in the body. Lean mass is everything else, muscle included.
In plain terms: three matched groups, one variable, eight months, and a scan before and after.
What the trial found
Fat mass fell in the two groups that did cardio and barely moved in the weights group. The group averages were −1.66 kg with cardio, −2.44 kg with both, and −0.26 kg with weights, which was not a significant change1. Cardio and both beat weights; cardio and both were not significantly different from each other.
Change in fat mass after 8 months, by programme · kg
Lean mass told the opposite story. It rose by 1.09 kg on average with weights and 0.81 kg with both, and did not change with cardio (−0.10 kg)1. A thigh scan agreed: more muscle area in the lifting groups.
Change in lean mass after 8 months, by programme · kg
Body weight followed fat and lean mass together. It fell in the cardio and combined groups and rose slightly in the weights group, because the lean mass they added outweighed the little fat they lost1. Body fat percentage still fell with weights, purely because lean mass went up.
The authors' conclusion was plain: cardio was "more effective than RT for the reduction of fat and body mass", the combined programme took double the time without significantly more fat loss, and a programme including weights is needed to add lean mass1.
In plain terms: cardio moved fat, weights moved muscle, and doing both moved both but cost twice the hours.
What 32 trials add
One trial is one trial. A 2021 network meta-analysis pooled 32 randomised trials, 4,774 adults with obesity, each lasting at least six months2. A network meta-analysis compares several options at once, using both direct head-to-head trials and indirect comparisons through a shared control group.
Against no exercise, fat mass fell by about 1.54 kg with cardio, 1.92 kg with combined training and 1.11 kg with weights (group averages, 95% confidence intervals in the chart)2. A confidence interval is the range the true average most likely sits in. The weights estimate was rated low certainty; cardio and combined were moderate.
Fat mass vs no exercise, by training type, across 32 trials · kg
Combined training ranked first for fat mass. Cardio ranked first for body weight and waist size. Weights and combined training both added about 1 kg of fat-free mass compared with cardio or no exercise; cardio did not change it2.
A separate 2022 review looked only at weights, in 54 trials of healthy adults. Compared with no exercise, lifting lowered body fat percentage by about 1.46 points and fat mass by about 0.55 kg, and it lowered visceral fat too (standardised effect −0.49)3. Small, but real. So "weights do nothing for fat" is too strong; "weights do less for fat than cardio, per week" fits the data.
In plain terms: all three work on fat. Cardio does more per hour, weights keep or add lean mass, and the two together look strongest for fat mass alone.
Why cardio moves fat faster
The simplest explanation is energy. Continuous cardio at a brisk pace uses more energy per minute than lifting, where much of the session is rest between sets. The 2021 review names that mechanism: aerobic work raises activity energy expenditure, and lifting changes it less2. That is the authors' reasoning, not something the trials measured directly.
Energy balance is also why exercise alone moves fat slowly. Neither trial group lost much in absolute terms over 8 months, because food intake stayed the same. How eating and moving combine is covered in calories in, calories out, and why the burn on a watch overstates it in why exercise burns fewer calories than the tracker says.
Common misreadings
- "Weights make you fat." No. The weights group gained weight on the scale because it gained lean mass; its fat mass did not rise1.
- "Cardio eats muscle." Not here. Lean mass did not change with cardio alone over 8 months1. Whether cardio slows muscle gain when added to lifting is a different question, covered in does cardio kill your gains.
- "Fat loss means scale loss." The scale mixes fat, lean mass and water. Fat mass and lean mass were measured separately for exactly this reason.
When eating less is part of the picture
Most people asking this question are also eating less. A 2026 review of 34 trials looked at that case: adults with overweight or obesity on calorie restriction, with or without exercise5. Adding exercise kept about 0.87 kg more fat-free mass on average; in the authors' words, exercise prevented nearly half (45.7%) of the fat-free mass loss. Mixed training showed the largest effect in the network analysis, then weights; cardio alone fell just short of statistical significance. A direct subgroup test found no significant difference between training modes.
Any deficit has limits. The calculator on this site caps it at 35% of TDEE and never sets a target below its calorie floor (1,500 kcal for men, 1,200 kcal for women on the standard setting); how big a calorie deficit should be explains both.
What the studies found
| Study | Who was studied | Key result | Year |
|---|---|---|---|
| Willis et al., *J Appl Physiol* (STRRIDE AT/RT)1 | 119 sedentary adults with overweight or obesity; 8 months, randomised | Fat mass: cardio −1.66 kg, both −2.44 kg, weights −0.26 kg (n.s.); lean mass up only with weights and both | 2012 |
| Morze et al., *Obes Rev*2 | 32 RCTs, 4,774 adults with obesity, ≥ 6 months | Fat mass vs no exercise: cardio −1.54 kg, combined −1.92 kg, weights −1.11 kg; combined ranked first for fat | 2021 |
| Wewege et al., *Sports Med*3 | 54 RCTs of healthy adults, weights vs no exercise | Body fat −1.46 percentage points; fat mass −0.55 kg | 2022 |
| Chen et al., *Obes Rev*4 | 84 RCTs, 4,836 people with overweight or obesity | Cardio, weights, combined and interval training all lowered visceral fat; weights ranked last | 2024 |
| Deller et al., *Diabetes Obes Metab*5 | 34 RCTs, 1,455 adults with overweight or obesity eating less | Exercise kept about 0.87 kg more fat-free mass than restriction alone | 2026 |
Limits worth knowing
- One trial at the centre. STRRIDE AT/RT analysed 119 of 196 randomised people: completers with consistent scans only, not everyone assigned1.
- Not equal time. Cardio was prescribed about 133 minutes a week and weights about 180, and "both" nearly doubled either. Per hour, cardio had the edge; per programme, the comparison is not perfectly matched.
- Population. Sedentary, middle-aged adults with overweight or obesity. Lean or trained people were not studied.
- Machines, not free weights. The weights programme used eight machines, three sets each.
- Inconsistency in the big review. The 2021 network meta-analysis found statistical inconsistency for every outcome, and certainty was low to moderate2.
- Abstract only for the reviews from 2022, 2024 and 2026; numbers here stay inside their abstracts.
Latest research
The newest study we opened is the 2026 review of calorie restriction with and without exercise5. It does not ask which mode loses more fat. It asks which keeps more lean tissue while eating less, and mixed training came out on top (+1.20 kg against restriction alone), with weights next. Read together with the 2021 review, the picture is consistent: cardio for fat, weights for lean mass, both if the time is there. A 2024 review of 84 trials on visceral fat, the fat around the organs, ranked vigorous cardio and interval training highest and weights lowest4.
The short version
- In an 8-month trial, cardio lost clearly more fat than weights; weights added lean mass instead.
- Across 32 longer trials, all three training types lowered fat mass; combined training ranked first for fat.
- Weights alone lower body fat a little, about half a kilogram of fat mass on average.
- Doing both took about double the time in the trial.
- One trial and a handful of reviews, not a prescription.
Educational overview only — not medical advice.
