Find Your Real TDEE From Your Food Log: Weight-Trend Math

Your most accurate TDEE is not a formula's output — it is the number your own food log and weight trend already contain. Average your logged intake over 2 to 4 weeks, convert your weight-trend change into energy at roughly 7,700 kcal (32,200 kJ) per kilogram — about 3,500 kcal per pound — and adjust your average intake by that daily imbalance. As of August 2026, no population formula matches a consistent personal log for accuracy, because your log measures your body directly.

Every TDEE calculator on the internet, including ours, hands you a population estimate. It is a starting point, and a genuinely useful one — but it was fitted to other people's bodies. If you have been tracking food and body weight for even a few weeks, you are holding better data than any equation: a record of what you actually ate, and what your weight actually did in response. This post walks through the arithmetic that turns those two columns into your real maintenance number, the same core idea behind "adaptive TDEE" features in modern tracking apps.

You may want our TDEE calculator in AU and US units for the starting estimate, and our review of why standard activity multipliers overestimate for the reasons a formula drifts. This piece covers what replaces them once your own data arrives. Every number below traces to a primary source listed at the end.

Why is a formula TDEE only a starting guess?

Formula TDEEs are population averages applied to an individual, and the errors compound: the resting-metabolism equation carries its own spread, the activity multiplier adds more, and neither updates as your body changes. A weight-trend TDEE replaces those assumptions with measured cause and effect — your intake in, your weight change out.

Static prediction rules perform poorly even in controlled research. When Thomas and colleagues tested the classic static model against measured outcomes, participants lost 20.1 lb (9.1 kg) where the 3,500-kcal rule predicted 27.6 lb (12.5 kg) — an overprediction of 7.4 lb (3.4 kg) in roughly nine weeks [4]. A validated dynamic model that accounted for body composition, age, height, and sex landed far closer, predicting 18.4 lb against an actual 17.0 lb in the same analysis [4]. The lesson is not that maths fails; it is that fixed constants applied to a moving body drift. We unpacked that history in why the 3,500-calorie rule is wrong.

Energy balance itself is also slower than most people assume. Hall's Lancet modelling work found that each sustained 100 kJ-per-day change in intake eventually shifts bodyweight by about 1 kg — but with a half-time of around one year [1]. Your body is always en route to an equilibrium it has not reached yet, which is why a TDEE measured from your own recent trend beats a formula frozen at day zero.

There is a practical upside hiding in that research: the weight-trend method does not care why your burn is what it is. Genetics, NEAT, an office job, a physical job, a GLP-1 prescription, years of dieting history — all of it is already priced into the relationship between your logged intake and your scale trend.

How do you calculate your real TDEE from your food log?

Real TDEE equals your average daily logged intake minus the energy value of your daily weight change. Log both for 2 to 4 weeks, convert the weekly weight trend to energy at 7,700 kcal (32,200 kJ) per kilogram or 3,500 kcal per pound, divide by seven, then add that daily figure to average intake if you are losing, or subtract it if you are gaining.

The energy-balance equation behind this is the same one Hall's group uses in research: change in body energy stores equals energy in minus energy out [1]. Rearranged for a tracker, it reads:

TDEE = average daily intake − (weekly weight change × energy per kg ÷ 7)

where weekly weight change is negative when you are losing. The 7,700 kcal per kg working constant is the research shorthand — formally 32.2 MJ per kilogram — that Hall's 2008 analysis traces through the clinical literature [3].

Here is the method in five steps:

1. Log intake honestly for 14 to 28 days. Everything counts — oils, coffee add-ins, weekend meals, the spoonful of peanut butter. Estimated entries are fine; skipped entries are what break the maths.
2. Weigh in daily, under the same conditions. First thing in the morning, after the bathroom, before food or water gives the cleanest signal.
3. Average each week's weigh-ins, then compare weekly averages. The difference between week-one average and week-three or week-four average, divided by the number of weeks between them, is your weekly trend.
4. Convert the trend to energy. Multiply the weekly change in kilograms by 7,700 kcal (32,200 kJ), or the change in pounds by 3,500 kcal, then divide by seven for a daily figure.
5. Adjust your average intake by that daily figure. Losing weight means your TDEE sits above your intake; gaining means it sits below.

A worked example in both unit systems:

What can throw the weight-trend method off?

Three things: water masquerading as tissue change, gaps in the log, and the energy constant itself. The first two are handled with longer windows and consistent logging; the third matters mainly if you are already lean, where 7,700 kcal per kg overstates the deficit each kilogram represents.

Water is the loudest noise source. Glycogen — your stored carbohydrate — is banked with three to four parts of water per part of glycogen, which is why the first week of any lower-carbohydrate or lower-calorie phase produces a scale drop far larger than its energy deficit explains [5]. Research on very-low-calorie diets found early rapid losses largely reflect glycogen and water mobilisation rather than fat [5]. The fix is patience: discard the first 7 to 14 days of a new phase from your calculation window, and let weekly averages — not single weigh-ins — define the trend.

Logging gaps are the second distortion, and they are close to universal. In the classic New England Journal of Medicine study of self-described "diet-resistant" adults, participants underreported actual food intake by an average of 47 percent and overreported physical activity by 51 percent — while their measured metabolisms sat within 5 percent of predicted values [2]. If that pattern sounds familiar, our breakdown of the 1,200-calories-and-not-losing maths covers where unlogged energy hides. The weight-trend method has a quietly forgiving property here: a consistent logging bias cancels out when you use the resulting TDEE as a target in the same app, with the same habits. Your computed number may sit below your true burn, but it is wrong by the same amount your future logs will be, so the planned deficit still lands. Inconsistent logging — strict weekdays, unlogged weekends — is what genuinely breaks the method.

The third caveat is the constant. Hall's analysis shows 7,700 kcal per kg approximately matches the energy density of lost weight for people with more than about 30 kg of body fat, but overestimates it for leaner people, because a larger share of their loss comes from lean tissue with a much lower energy density [3]. In practice:

How often is it worth recalculating?

Every 4 to 6 weeks during an active fat-loss or gaining phase, or after roughly each 2 to 3 kg (4 to 7 lb) of change — because the number you measured is attached to the body that produced it.

TDEE is not a fixed trait. As weight falls, a smaller body burns less, and adaptive responses trim expenditure further — the dynamics that make weight loss curvilinear rather than linear in validated models [1][4]. The NIDDK's Body Weight Planner exists precisely because static targets fail over long timelines; it lets users build calorie plans that update for a goal weight and timeframe rather than assuming one deficit works forever [6]. Your log gives you the same power with even better inputs: rerun the five steps on your most recent 3 to 4 weeks whenever progress stalls, and treat each result as the current reading, not a permanent setting.

One practical note for KCALM users and anyone else who photographs or logs meals: the method only requires the two data streams you are already producing. There is no extra protocol, no lab visit, and no dependence on a wearable's burn estimate — helpful, given how loosely watch calories track measured expenditure, as we covered in whether to eat back exercise calories.

Frequently Asked Questions

How long do I need to track before the number is reliable?

Two weeks is the workable minimum; three to four is meaningfully better. Shorter windows are dominated by water and gut-content noise, which can swing the scale 0.5 to 2 kg between mornings. Weekly averages across a 21-to-28-day window smooth most of it out, and excluding the first week of any new eating phase removes the glycogen-driven drop [5].

Does the method work if I underreport calories?

Surprisingly well, provided the underreporting is consistent. If your log runs 15 percent light every day, the TDEE you compute is 15 percent light too — and because you will keep logging the same way, targets set from that number still produce the planned trend. Research suggests underreporting averaged 47 percent in self-described diet-resistant adults [2], so the bigger risk is uneven logging: strict Monday-to-Friday entries with untracked weekends distort the trend rather than merely shifting it.

Is 7,700 kcal per kilogram accurate for everyone?

It is a working constant, not a law. Hall's modelling shows it approximates the energy density of lost weight when initial body fat exceeds roughly 30 kg, and overstates it for leaner people, whose losses include proportionally more lean tissue [3]. Treat your computed TDEE as a tight range — plus or minus 100 kcal (420 kJ) or so — and let the next month of data refine it.

What if my weight jumps around from day to day?

That is expected, and it is why the method uses weekly averages instead of single readings. Glycogen holds three to four times its weight in water [5], sodium and gut contents move the scale daily, and none of it is fat. Daily weigh-ins feel noisier but produce a smoother average than sporadic ones — more data points, better trend.

Can I use this method while trying to gain muscle?

Yes — the arithmetic is symmetric. A gaining trend converts to energy the same way, and the result is your intake minus the daily surplus. The constant is less precise for gains, since new tissue is a mix of muscle, glycogen, and fat with a different energy density than lost tissue, so lean-gain phases benefit from the same plus-or-minus-100-kcal humility and regular recalculation.

Do I still need a formula TDEE at all?

For the first two to four weeks, yes — a formula is the sensible starting point before your data exists, and our Mifflin-St Jeor versus Katch-McArdle comparison covers picking one. After that, the formula has done its job: your log takes over, and every recalculation makes the number more yours.

Sources

1. Hall KD, Sacks G, Chandramohan D, et al. Quantification of the effect of energy imbalance on bodyweight. The Lancet. 2011. https://pmc.ncbi.nlm.nih.gov/articles/PMC3880593/
2. Lichtman SW, Pisarska K, Berman ER, et al. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. New England Journal of Medicine. 1992. https://pubmed.ncbi.nlm.nih.gov/1454084/
3. Hall KD. What is the required energy deficit per unit weight loss? International Journal of Obesity. 2008. https://pubmed.ncbi.nlm.nih.gov/17848938/
4. Thomas DM, Martin CK, Lettieri S, et al. Can a weight loss of one pound a week be achieved with a 3500-kcal deficit? Commentary on a commonly accepted rule. International Journal of Obesity. 2014. https://pmc.ncbi.nlm.nih.gov/articles/PMC4024447/
5. Kreitzman SN, Coxon AY, Szaz KF. Glycogen storage: illusions of easy weight loss, excessive weight regain, and distortions in estimates of body composition. American Journal of Clinical Nutrition. 1992. https://pubmed.ncbi.nlm.nih.gov/1615908/
6. National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Body Weight Planner. https://www.niddk.nih.gov/health-information/weight-management/body-weight-planner

This article is general information, not medical advice. Energy needs vary between individuals — consider speaking with your GP or an Accredited Practising Dietitian before making significant changes, particularly if you have a medical condition or a history of disordered eating.