Methodology version 2.0

How Macro Targets calculates and calibrates your plan

Published: July 14, 2026 · Last reviewed: July 14, 2026

Macro Targets is an educational planning tool for generally healthy adults. This page documents the exact calculation sequence, the evidence used to shape it, what the tool intentionally excludes, and what remains uncertain.

1. Intended population and safety exclusions

The calculator is intended for adults age 18 and older who want a general starting estimate for weight loss, maintenance, or gradual weight gain. It is not designed for:

• People under age 18.
• Pregnancy or breastfeeding.
• Current or recent eating-disorder treatment.
• A clinician-directed diet or a medical condition or medication change that materially affects weight, appetite, fluid balance, or energy needs.
• A weight-loss goal below a BMI of 18.5 when height is available to the standard-formula mode.

For those situations, the general calculator stops or asks the user to work with a qualified clinician. These exclusions are directionally consistent with the adult-use limits published by the NIDDK Body Weight Planner.

2. Maintenance calorie estimate

Standard formula

The standard formula uses the Mifflin-St Jeor equation to estimate resting energy expenditure (REE), followed by an activity multiplier.

Activity categories are broad and are often the largest source of error in equation-based estimates. The result should be calibrated from real intake and weight trend rather than treated as a measurement.

Wearable estimate

The tool labels this a wearable estimate, not measured mode. Consumer devices can be systematically high or low for energy expenditure. The input confidence label reflects only the number of days averaged:

• Low: 1-3 days.
• Medium: 7-14 days.
• Better starting estimate: 21-28 days.
• Calibrated: sufficient logged intake and weight-trend data in the adaptive check-in.

3. Goal calories and pace guardrails

The 500 kcal conversion is used only as a short-horizon planning approximation. It does not model long-term metabolic adaptation.

• Weight-loss pace cap: the lower of 1% of current body weight per week or the pace that would reach the tool calorie floor.
• Weight-gain pace cap: the greater of 0.25 lb/week or 0.25% of current body weight per week.
• Tool calorie floor: 1,200 kcal/day when the female equation is selected and 1,500 kcal/day when the male equation is selected.

These floors are product guardrails, not individualized prescriptions or proof that an intake above the floor is appropriate. Aggressive requests are reduced, and targets at or below the floor are not produced for weight loss.

4. Macro target logic

Macro Targets uses a sequence rather than a fixed calorie percentage:

1. Choose a protein factor from training type.
2. Add a small goal and preference adjustment, capped at 2.2 g/kg/day.
3. Use goal weight as the protein reference during weight loss; otherwise use current weight.
4. Reserve a minimum fat target of 0.6 g/kg/day or 35 g/day, whichever is higher.
5. Allocate remaining calories to carbohydrates, preserving at least 50 g/day in the calculation (100 g/day for endurance-focused training).

Weight loss adds 0.2 g/kg/day. A protein-forward preference adds 0.2 g/kg/day; lower carb adds 0.1 g/kg/day. The displayed protein range is generally the working factor plus or minus 0.2 g/kg/day, subject to available calories and the fat/carbohydrate guardrails.

The result emphasizes calories and a protein minimum. Exact carbohydrate and fat grams are flexible and may be moved within the same calorie budget.

5. Pace-based goal window

The central projection divides the remaining weight change by the selected weekly pace. The displayed window uses 115% of that pace for the optimistic date and 85% for the conservative date.

This is explicitly a pace-based projection, not a dynamic physiological model. Longer-term changes often slow as body weight, expenditure, adherence, and adaptation change. For a more detailed dynamic model, use the NIDDK Body Weight Planner.

6. Adaptive check-in

The adaptive check-in requires at least 7 valid date-and-weight rows spanning at least 7 calendar days. Fourteen to 21 days is preferred.

1. Convert weights to pounds internally when necessary.
2. Fit a least-squares linear trend across all valid weigh-ins.
3. Convert the daily slope to an actual weekly pace.
4. Compare the signed actual pace with the signed target pace.
5. Treat a difference within 0.15 lb/week or 25% of target pace, whichever is larger, as close enough to hold steady.
6. Otherwise convert the pace difference using 500 kcal per lb/week, round to 50 kcal, and cap the adjustment at plus or minus 200 kcal/day.
7. Apply the adjustment to average logged intake when it is provided, not blindly to the original target.

Protein and fat targets are held steady while carbohydrates absorb the calorie change first. The check-in never returns a revised target below 1,200 kcal/day, but that browser-side floor is still not individualized medical guidance.

7. Confidence rating

• Low: the entries cover too little time for an adjustment.
• Medium: at least 7 weigh-ins across at least 7 days.
• Better starting estimate: at least 14 weigh-ins across at least 14 days with logged intake.
• Calibrated: at least 18 weigh-ins across at least 18 days, logged intake, and trend residual variation no greater than 1 lb.

The rating is a product heuristic. It is not a statistical confidence interval and does not detect inaccurate food logging, fluid shifts, illness, menstrual-cycle effects, or changes in sodium, glycogen, training, medication, or activity.

8. Privacy and analytics boundaries

• The initial calculator is submitted to and calculated on the web server. Values are not intentionally written to a user profile or calculator database.
• Form preferences and saved plans can be stored in local storage on the user's device.
• Adaptive check-in parsing, trend analysis, and recommendations run in the browser. Check-in data is not submitted with a form.
• Analytics events use names such as calculation_completed, calibration_started, and checkin_completed. Weight, age, sex, calorie, goal, and macro values are not included in those events.
• Exported JSON files are created locally and are controlled by the user.

9. Primary references

1. Mifflin MD, St Jeor ST, et al. A new predictive equation for resting energy expenditure in healthy individuals. American Journal of Clinical Nutrition. 1990;51(2):241-247.
2. Morton RW, Murphy KT, et al. A systematic review, meta-analysis and meta-regression of protein supplementation and resistance training outcomes. British Journal of Sports Medicine. 2018;52:376-384.
3. Fuller D, Colwell E, et al. Reliability and validity of commercially available wearable devices. JMIR mHealth and uHealth. 2020;8(9):e18694.
4. National Institute of Diabetes and Digestive and Kidney Diseases. Body Weight Planner. Accessed July 14, 2026.

10. Known limitations

• Mifflin-St Jeor and activity multipliers estimate population averages, not individual metabolism.
• Wearable energy estimates vary by device, algorithm, activity, and user.
• The short-horizon calorie-to-weight approximation does not model long-term adaptation.
• Protein factors are planning ranges and do not account for every clinical, dietary, or body-composition context.
• Linear weight trends can be distorted by short data spans or systematic changes in hydration and logging.
• No independent clinical review has been completed yet.

11. Revision history

Contact

Questions or corrections can be sent to help@macrotargets.com.