How Much Muscle Should I Have? | Tommy Wood

Wood explains that FFMI was chosen over ALMI in his population work precisely because it requires no imaging — you can compute it with any body-fat estimate and a scale. The target ranges he recommends (FFMI 16-17 for women, 19-20 for men) give a buffer above the risk cutoffs (14/17) to account for normal measurement error, individual variation, and the inevitable age-related decline that makes building a reserve now critical. Once above the protective threshold, the evidence for 'more is better' weakens — a FFMI of 22 being optimal is likely a data artefact.

Wood argues that the public health system and wellness media have been so fixated on obesity that the sarcopenia crisis has been entirely overlooked. You can be normal weight on BMI and have dangerously low FFMI. Conversely, a heavily muscled person may register as 'overweight' on BMI with no elevated health risk. The practical protocol is to make FFMI the north-star metric: set a target FFMI in the 17-20 range (women/men respectively), and design nutrition and training to move toward it regardless of what the scale says.

Wood frames this carefully: the message 'just lift heavy like I do' coming from a heavily muscled person alienates most listeners. The clinical reality is that any movement to stimulate muscle tissue in a sedentary person has multiple beneficial downstream effects — and the improvement in muscle quality (contractile function, fibre type distribution, mitochondrial density) can be meaningful even before visible changes in FFMI occur. He is explicit: 'it doesn't take much right any movement to stimulate that muscle tissue particularly in people who are sedentary has multiple beneficial downstream effects.'

Wood tells a story about a GP at the British Society of Lifestyle Medicine conference (the second-largest lifestyle medicine society in the world) who had been telling his wife to lift weights for years with no result. After she heard Wood speaking on Rangan Chatterjee's podcast, she sent her husband the episode and said 'look — here's a woman talking about muscle mass, now I think I should lift weights.' Wood's conclusion: 'you have a sphere of influence of people who are more like you who are maybe more likely to listen to you than they are to listen to me.' The principle generalises — post-menopausal women respond to post-menopausal women, sedentary office workers respond to other sedentary office workers who transformed.

Wood notes the scientific tension: for mechanistic research you want to isolate exactly skeletal muscle tissue (hence interest in CT, MRI, and D3-creatine methods), but for practical population guidance, FFMI delivers essentially the same information at a fraction of the cost and complexity. He cites William Evans's D3-creatine work as the leading edge of more specific measurement, but for today's clinical and self-monitoring use, FFMI is the right tool.

Tommy Wood explains: FFMI is calculated the same way as BMI — divide fat-free mass in kilograms by height in metres squared. You just need a body-fat estimate (InBody, DEXA, or even a skinfold) and your weight. ALMI, by contrast, isolates just the skeletal muscle in your arms and legs, which is closer to the metabolically relevant mass but requires a DEXA and extra calculation. Wood's own research compared both and found the relationship between low muscle mass and cognitive function was essentially identical whether you used ALMI or FFMI — suggesting that for practical guidance, the simpler metric is sufficient. The upshot: stop talking only about obesity percentages and start tracking FFMI.

Wood explains that the cutoffs differ somewhat between studies — which is why he recommends targeting above the risk threshold rather than sitting on it. His pragmatic recommendations: women should aim for FFMI 16-17, men 19-20. He notes that the 'sweet spot' is well below the FFMI 25 ceiling associated with peak natural muscularity — the goal for most people is not to maximise muscle but to stay clear of the low-muscle danger zone. The frame is protective, not aesthetic: you need enough functional muscle tissue that losing some to illness, inactivity, or ageing does not immediately push you into the high-risk range.

Wood acknowledges that getting to FFMI 22 would be great — but says recommending that as a population target misunderstands the data. Men and women have systematically different lean mass relative to height, so treating a single FFMI cutoff as sex-neutral is biologically implausible. The more reliable interpretation is that the mortality curve flattens out somewhere around 19-20 for men and 16-17 for women — enough protective muscle mass that more is unlikely to confer additional survival benefit for a general population, even if athletes perform better at higher levels.

Wood describes studies looking at brain atrophy (how much the brain shrinks with age) and elements of cognitive function, and finding that muscle mass outperforms both fat mass and BMI as a predictor. In his own work, he tested both ALMI and FFMI and found the association with cognitive outcomes was essentially identical — which also validated FFMI as a simpler proxy. The muscle-brain link likely operates through multiple mechanisms: myokines secreted by contracting muscle (including BDNF precursors), systemic metabolic regulation (insulin sensitivity, glucose clearance), and possibly direct neurovascular effects. Wood frames muscle as a secretory organ and a whole organ system, not just contractile tissue.

Wood frames this as the key practical takeaway: 'I don't care about the mass that you exert on a scale — what I care about is how much functional lean mass you have, and that can improve and increase regardless of anything else.' A sedentary person can increase FFMI through resistance training without losing a single kilogram of total weight. A person on a crash diet may decrease FFMI even as they lose total weight. The weight-neutral nature of FFMI makes it a more psychologically productive target, especially for populations that have been harmed by weight-focused messaging.