Adaptive TDEE from your own logs

Every formula is wrong about you specifically

Any calorie calculator you've ever used starts from a population equation. Population equations carry real individual error. Two people with identical stats can differ by hundreds of calories a day. The honest approach isn't a better first guess; it's treating the first guess as temporary and replacing it with your data.

Step 1: A documented starting point

Onboarding uses the Mifflin-St Jeor equation, the standard clinical choice for resting energy expenditure, multiplied by a conventional activity factor, then adjusted for your goal. It's labeled what it is: an initial estimate with meaningful uncertainty.

Step 2: Energy balance takes over

Once you log, the arithmetic is old-fashioned energy balance: over a multi-week window, expenditure equals confirmed intake minus the energy stored or released by your weight change. If you averaged 2,600 kcal and gained slowly, your true maintenance sits near the math, not near the onboarding guess. Kevin Hall's work is the backbone here, both for why the energy-per-unit-weight conversion is an approximation, and whydense weigh-ins over multi-week windows are required before the estimate tightens.

The part most trackers get wrong: honest inputs

Adaptive TDEE breaks silently when incomplete logging days pollute the intake average. Flexbound's fix is structural: you mark each day'scompleteness, and only complete days teach the maintenance estimate. A day you half-logged never drags your computed TDEE downward.

• Complete days → feed the energy-balance window
• Partial days → kept as notes, excluded from the math
• Each estimate shows its source, serving assumption, and confidence

Goal targets stay conservative

Deficits and surpluses are set as fractions of adapted maintenance following evidence-based body-composition guidance. Sustainable rates that preserve lean mass, not crash targets. And because the estimate is deterministic, when it changes, the app can tell you exactly why: which weeks, which weigh-ins, which arithmetic.

Limits, stated plainly

Short windows, sparse weigh-ins, and water-weight noise all widen the error bars. That is why confidence is graded and shown rather than hidden. Large day-to-day fluctuations early in a diet phase are mostly water and glycogen; the multi-week window exists precisely to see through them.