The Core Answer: How to Calculate Your Macros for a Keto Diet
If you want to know how to calculate your macros for a keto diet without a black-box tool, the process is three steps: estimate basal metabolic rate (BMR), adjust for activity to get total daily energy expenditure (TDEE), then split those calories into keto ratios. A standard therapeutic keto split is roughly 70–75% fat, 20–25% protein, and 5–10% net carbs. Convert percentages to grams using 9 calories per gram of fat and 4 per gram of protein or carb.
When I first calculated keto macros by hand for a 48-year-old client with a sedentary job and borderline A1C, I made the mistake of applying a flat 75% fat ratio to a 2,000-calorie TDEE without checking her lean mass. She ended up with 41g of protein—too low for her muscle mass—and stalled after six weeks. That experience taught me to anchor protein to body composition, not just percentages.
For a quick manual example: a 35-year-old male, 180 lb (81.6 kg), 5’10” (177.8 cm), lightly active. Mifflin-St Jeor BMR = 10×81.6 + 6.25×177.8 – 5×35 + 5 = 1,782 kcal. TDEE ×1.375 = 2,450 kcal. At 75% fat, 20% protein, 5% carbs: fat = 204g, protein = 122g, carbs = 31g net. We’ll refine this later.
The takeaway: manual calculation is not just arithmetic; it’s a clinical reasoning exercise. If you prefer automation, our Keto Macro Calculator mirrors these formulas, but understanding the bones prevents tool errors.
Step 1: Calculate BMR With the Mifflin-St Jeor Equation
The most reliable pen-and-paper formula is the Mifflin-St Jeor equation, validated in clinical nutrition settings. For men: BMR = 10×weight(kg) + 6.25×height(cm) – 5×age + 5. For women: the final term is – 161 instead of +5.
Why Mifflin-St Jeor Beats Older Formulas
The legacy Harris-Benedict equation overestimates BMR by 5–10% in modern sedentary populations, according to comparative studies. In my practice, that gap translates to a 100–150 kcal daily surplus if you use Harris-Benedict blindly—enough to blunt fat loss on keto.
Convert pounds to kilograms (÷2.2) and inches to centimeters (×2.54) before plugging in. A common error: people use total body weight even when obese, which inflates BMR. If body fat exceeds 30%, I subtract an adjusted fat mass estimate or use lean mass as a cross-check.
When to Use Katch-McArdle Instead
If you know your body fat percentage from a DEXA scan, the Katch-McArdle formula (BMR = 370 + 21.6 × lean mass kg) is even tighter. I use it for athletes because it removes the fat-mass penalty. For a 90kg man at 15% body fat, lean mass = 76.5kg, BMR ≈ 2,023 kcal—noticeably higher than Mifflin’s 1,850.
The thing nobody tells you about BMR formulas: they assume thermoneutral conditions. Real life—cold offices, fidgeting—adds variance. Treat the number as a floor, then let hunger and scale weight adjust it.
Step 2: Multiply by Activity to Get TDEE
BMR is what you burn at rest. To get TDEE, multiply by an activity factor: 1.2 (sedentary), 1.375 (light), 1.55 (moderate), 1.725 (heavy), 1.9 (athlete). The thing nobody tells you about these multipliers is that most people overestimate their own category by one full level.
I once had a remote client who claimed “moderate” training five days a week, but his logs showed 20-minute walks. We corrected to 1.2, dropping his TDEE from 2,900 to 2,300 kcal. His previous macro plan had him gaining weight on keto—a confusing outcome until we fixed the input.
Non-exercise activity thermogenesis (NEAT) matters. A waiter on feet 8 hours daily may need a 1.4 multiplier despite no “workouts.” I ask clients to wear a pedometer; under 5,000 steps defaults to sedentary regardless of gym sessions.
If you want a sanity check, use a heart-rate tracker for a week. But even then, consumer devices can misreport by 10–20%. Treat TDEE as a starting hypothesis, not gospel.
Step 3: Convert Calories Into Keto Macro Grams
Now allocate TDEE across macros. Two practitioner approaches exist: percent-calorie ratio and gram-per-kg lean body mass. The first is simpler; the second is precise for clinical populations.
Percent-Calorie Method
Take TDEE × target %. Fat: calories÷9. Protein: calories÷4. Carbs: calories÷4. Example: 2,450 kcal × 75% = 1,837 fat cal → 204g fat. This matches many online tools, including our Keto Macro Calculator, which automates the math if you skip the worksheet.
The trap: at low calorie levels, 20% protein can mean 40g—insufficient for a 70kg person. I’ve seen women on 1,200 kcal diets get only 60g protein, triggering hair loss. Percentages scale down dangerously when calories drop.
Gram-Per-Lean-Mass Method
For diabetics or thyroid patients, I set protein at 1.2–1.6g per kg of lean body mass, carbs at 20–30g net, and fill remaining calories with fat. This prevents the protein underfeeding I saw earlier. A 90kg man with 70% lean mass gets ~75–95g protein minimum, not the 122g from pure ratios.
Below is a comparison of the two frameworks I use in clinic:
| Approach | Best For | Protein Rule | Primary Risk |
|---|---|---|---|
| Percent-Calorie | General beginners | 20–25% of kcal | Underfeeds lean mass at low kcal |
| Gram-per-Lean-Mass | Diabetics, Hashimoto’s, athletes | 1.2–1.6g/kg LBM | Requires body comp scale |
Pick the method that matches your health context, not the one that’s easiest. The manual worksheet later blends both.
What Is the 2-2-2-2 Rule on Keto—and Why It’s Usually Flawed
The 2-2-2-2 rule circulating in forums is a memory aid suggesting daily caps of roughly 20g net carbs, 20g protein, 20g fat, and sometimes 20g fiber or 2 liters of water. It’s meant to simplify keto for newcomers who fear calculation. In reality, the rule is a crude starting point that ignores individual energy needs.
Some variations replace the fourth “2” with two meals or two cups of greens, but the constant is the 20g net carb ceiling. I tested the literal numbers on a 110kg male athlete; his 20g protein limit was less than half his maintenance need, causing muscle loss within a month. The rule’s only defensible use is as a strict induction-phase carb cap, not a full macro prescription.
Most people don’t realize that 20g fat is only 180 calories—a snack, not a day. If you follow the literal 2-2-2-2 numbers, you’ll be in a severe calorie deficit unless you eat many times those amounts. Use it as a cautionary tale, not a template.
The flawed logic stems from conflating “ketogenic threshold” (usually 20–50g net carbs) with total intake. The threshold is real; the uniform protein and fat numbers are not. That distinction clears up why some newcomers feel great for a week then crash.
Adjusting Macros for Specific Goals: Loss, Maintenance, Performance
Once you have a baseline TDEE, tweak the split. For fat loss, create a 15–20% deficit (multiply TDEE by 0.8–0.85) but keep protein at the higher end (1.6g/kg LBM) to spare muscle. For maintenance, eat at TDEE with 70% fat, 25% protein, 5% carbs.
Fat Loss: The Deficit That Preserves Ketosis
I counsel clients to never drop below 1,200 kcal total on keto; below that, adrenal stress rises. A 2,000 kcal TDEE woman uses 1,700 kcal target, 30g net carbs, 95g protein, remaining 132g fat. That’s sustainable for months.
Maintenance and Reverse Dieting
After a cut, add 50–100 kcal weekly from fat only, keeping carbs fixed. This prevents rebound glucose spikes. I’ve used this with a 54-year-old who kept A1C at 5.4% post-loss.
Performance and Endurance Keto
Athletes often need 70–80g net carbs (targeted keto) around training, pushing carb % to 15–20% while trimming fat. I’ve seen cyclists crash on 20g carbs during 3-hour rides; a 30g pre-ride dose fixed it. Use the matrix below.
| Goal | Calorie Multiplier | Net Carbs | Protein g/kg LBM | Fat % |
|---|---|---|---|---|
| Loss | 0.8–0.85 TDEE | 20–30g | 1.6 | 70–75% |
| Maintenance | 1.0 TDEE | 30–50g | 1.2–1.4 | 70% |
| Performance | 1.0–1.1 TDEE | 50–80g | 1.6–2.0 | 60–65% |
Notice fat % falls as carbs rise; calories stay balanced. This matrix is the exact decision tool I hand to clients after the worksheet.
Keto Macros for Diabetics: Will a Keto Diet Lower My A1C?
Will a keto diet lower my A1C? For many with type 2 diabetes, carbohydrate restriction reliably reduces post-meal glucose spikes, and several clinical cohorts show A1C drops of 0.5–1.0% over 12 weeks. The National Institute of Diabetes and Digestive and Kidney Diseases notes that carb intake is the strongest dietary lever for blood glucose control.
However, macros still matter. I advise diabetic clients to keep net carbs under 30g and protein moderate (1.2g/kg LBM) because excessive protein can trigger gluconeogenesis. Fat should come from unsaturated sources, not just butter. The trade-off: rapid A1C improvement may require medication adjustments under a doctor’s care to avoid hypoglycemia.
One edge case: lean type 1 diabetics still need sufficient fat to prevent hypoglycemia during overnight fasts. The manual calculation must preserve at least 1,200 kcal from fat for safety. Never treat keto as a substitute for medical supervision.
In my clinic, I recheck fasting insulin at 6 weeks. If A1C falls but LDL rises sharply, we shift fat from saturated to monounsaturated. That nuance rarely appears in calculator outputs but is critical for long-term vascular health.
Can Hashimoto’s Patients Go on Keto? Protein and Fat Moderation
Can Hashimoto’s patients go on the keto diet? Yes, but with caveats. The autoimmune thyroid condition often coincides with slowed metabolism, so using a raw TDEE from standard activity charts may overestimate needs by 10–15%. The NIDDK highlights that thyroid hormone levels dictate energy expenditure.
In my experience, Hashimoto’s clients do best on a modified keto: 60–65% fat (not 75%), 25–30% protein, and 30–40g net carbs from cruciferous-free vegetables if they have GI sensitivity. High saturated fat loads can worsen inflammatory markers in some; we use olive oil and avocado instead of heavy cream.
Most people don’t realize that excessive protein restriction on keto can depress T3 conversion. I keep protein at 1.2g/kg LBM minimum. If TSH rises after 8 weeks, we add 10–15g carbs from berries or squash. This is a negotiation, not a rigid rule.
Another blind spot: selenium and iodine intake must be stable before starting. I request a baseline thyroid panel; if antibodies exceed 300 IU/mL, we introduce keto at 40g carbs to avoid adrenal crossover stress. The macro math is secondary to endocrine stability.
A Manual Keto Macro Worksheet You Can Use Today
Here is the exact worksheet I hand to clients. Fill each line with a pencil—not a phone app—to engage the math:
- Weight (kg) ___ ; Height (cm) ___ ; Age ___ ; Sex ___
- BMR = 10×wt + 6.25×ht – 5×age + (5 or –161) = ___ kcal
- Activity factor ___ → TDEE = BMR × factor = ___ kcal
- Goal multiplier (0.8 loss, 1.0 maint, 1.1 perf) = ___ kcal target
- Protein: 1.2–1.6g × kg lean mass = ___ g (×4 = ___ kcal)
- Net carbs: 20–50g = ___ g (×4 = ___ kcal)
- Fat: remaining kcal ÷9 = ___ g
Example fill for the 35-year-old male earlier: BMR 1,782; TDEE 2,450; goal 1.0; lean mass assumed 75kg ×1.4 = 105g protein (420 kcal); carbs 30g (120 kcal); fat = (2450-540)/9 = 212g. Slightly different from pure ratio but safer.
Validate with a food scale for two weeks. The scale doesn’t lie; apps do.
If the numbers feel off, repeat the BMR with Katch-McArdle. The worksheet is a loop, not a one-time sheet.
Common Mistakes and Edge Cases I’ve Seen in Practice
Beyond the 2-2-2-2 myth, the biggest error is treating “net carbs” as total carbs minus only fiber. Sugar alcohols vary; erythritol is mostly inert, but maltitol spikes glucose. I’ve watched a client’s ketone levels drop after sugar-free candy because they miscounted.
Another blind spot: women over 50 often need a 0.9 TDEE multiplier due to hormonal shifts. A blanket activity chart fails them. Also, if you train fasted, add 5–10g carbs post-workout to protect cortisol—another nuance calculators miss.
Electrolyte mismatch is the silent killer of homemade keto plans. At 200g fat daily, sodium needs hit 5,000mg; most clients get 3,000. I add 1 tsp salt to broth and track magnesium citrate separately. Lab tests every 6–8 weeks should override the spreadsheet.
Finally, remember that keto macro math is a model, not metabolism. Fasting insulin, A1C, and TSH are the real scoreboard. That’s the practitioner’s loop I trust more than any online form—and the reason I still calculate by hand first.