# Nutrition: KiteSink fitness field guide

Eating to support training, and tracking it honestly. Each entry gives the point, the reasoning, what to do, how well supported it is, and its sources.

## Guidance

### Body weight follows energy balance, but it moves slowly

**Eat above what you burn and weight rises, eat below and it falls, but the change takes months to show its full size.**

A mathematical model of adult metabolism found that body weight responds slowly to a change in intake, with a half time of about one year. It also estimated that a persistent gap of only about 30 kJ (7 kcal) a day lies behind the average weight gain seen in the US. In a ward study of 20 adults, weight change tracked energy intake closely (r = 0.8). Small daily gaps add up, and a single day rarely shows.

- Judge a calorie target over weeks, not days.

Energy and maintenance · Beginner · Some support · Sources: [Hall 2011](https://doi.org/10.1016/S0140-6736(11)60812-X), [Hall 2019](https://doi.org/10.1016/j.cmet.2019.05.008)

### Calorie calculators give a starting point, not your maintenance

**Even the best resting-rate equation is within 10 percent of measured for most people, and many individuals fall outside it.**

A systematic review compared four equations for resting metabolic rate. Mifflin-St Jeor predicted within 10 percent of the measured value for more people than the others, and it still missed for a share of individuals. Total daily need is that resting figure multiplied by an activity level, and the FAO/WHO/UNU report gives about 1.40 to 2.40 as the range sustained by free-living adults. A wide range of activity levels means a wide range of totals.

- Eat your estimate for two to three weeks while weighing yourself daily.
- Adjust by 100 to 200 kcal if weight moves the wrong way.

Energy and maintenance · Beginner · Well supported · Sources: [Frankenfield 2005](https://doi.org/10.1016/j.jada.2005.02.005), [FAO/WHO/UNU 2004](https://www.fao.org/4/y5686e/y5686e07.htm)

### Most lifters do well on 1.6 to 2.2 g of protein per kg a day

**About 1.6 to 2.2 g per kg (0.7 to 1.0 g per lb) of body weight a day covers most people who lift to build muscle.**

The ISSN position stand calls 1.4 to 2.0 g/kg a day sufficient for most exercising people. A meta-analysis of 74 trials found added protein helped lean mass in adults under 65 who ate 1.6 g/kg or more. A dose-response review of 105 trials (5,402 participants) found each extra 0.1 g/kg added 0.39 kg of lean mass below 1.3 g/kg a day and 0.12 kg above it. Returns shrink as intake rises.

- Set a daily target in grams from your body weight, then check it weekly.
- Take the lower half of the range if food budget is tight.

Protein · Beginner · Well supported · Sources: [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8), [Nunes 2022b](https://doi.org/10.1002/jcsm.12922), [Tagawa 2021](https://doi.org/10.1093/nutrit/nuaa104), [Iraki 2019](https://doi.org/10.3390/sports7070154)

### High protein does not harm kidney function in healthy adults

**In healthy adults without kidney disease, higher protein intakes did not reduce kidney filtration in trials.**

A meta-analysis pooled 28 trials with 1,358 participants who ate 1.5 g/kg a day or more, or at least 20 percent of energy, against lower intakes. Filtration rate was slightly higher after high-protein eating, and the change from baseline did not differ between groups. The trials were short and had some risk of selection bias. This does not cover people who already have kidney disease. They should ask their clinician.

- If you have kidney disease or a family history, check your protein plan with a clinician first.

Protein · Beginner · Well supported · Sources: [Devries 2018](https://doi.org/10.1093/jn/nxy197)

### Spread protein over several meals, but the split matters less than the total

**About 0.25 to 0.55 g per kg per meal is a sound target (20 to 40 g for most people), and the daily total is what matters most.**

The ISSN stand suggests 0.25 g/kg or 20 to 40 g per serving, spread across the day. A bodybuilding review pairs 1.6 to 2.2 g/kg a day with 0.40 to 0.55 g/kg per meal over 3 to 6 meals. Muscle protein synthesis levelled off at about 0.24 g/kg in younger men and 0.40 g/kg in older men. In eight adults, even meals raised 24-hour synthesis 25 percent more than a skewed pattern. A review calls the distribution evidence limited and inconsistent, and a tracer study found 100 g produced a larger, longer response than 25 g, so there is no hard cap per meal.

- Aim for three to five protein-containing meals.
- Do not worry about one bigger protein meal.

Protein · Intermediate · Some support · Sources: [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8), [Moore 2015](https://doi.org/10.1093/gerona/glu103), [Mamerow 2014](https://doi.org/10.3945/jn.113.185280), [Hudson 2020](https://doi.org/10.3390/nu12051441), [Trommelen 2023](https://doi.org/10.1016/j.xcrm.2023.101324), [Iraki 2019](https://doi.org/10.3390/sports7070154)

### Protein timing around training matters little next to the daily total

**Eating protein straight after training is fine, but a precise window does not change results once daily intake is met.**

A systematic review of 65 studies with 2,907 participants found protein supplementation raised lean body mass in adults and older adults, and subgroup analyses showed no benefit from any specific timing. The ISSN position says the anabolic effect of exercise lasts at least 24 hours. A protein-containing meal before or after training is a reasonable habit, and it is not urgent.

- Eat a protein meal within a few hours either side of training.

Timing, fluids and alcohol · Beginner · Well supported · Sources: [Wirth 2020](https://doi.org/10.1093/jn/nxaa049), [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8)

### Plant protein builds muscle about as well, given enough total protein

**When total protein is adequate, plant and animal sources give similar muscle and strength gains over months.**

A meta-analysis of 18 studies found protein source did not change absolute lean mass or strength, though animal protein favoured percent lean mass, and younger adults gained 0.41 kg more lean mass with animal protein. In a 12-week trial, 19 vegans and 19 omnivores on 1.6 g/kg gained muscle alike. A soy-versus-animal meta-analysis of nine studies agreed. Single-meal studies do show a lower response to soy and wheat, which is why total intake matters.

- If you eat no animal foods, aim for the upper half of the protein range.
- Mix soy, legumes and grains across the day.

Protein · Intermediate · Well supported · Sources: [Lim 2021](https://doi.org/10.3390/nu13020661), [Hevia-Larraín 2021](https://doi.org/10.1007/s40279-021-01434-9), [Messina 2018](https://doi.org/10.1123/ijsnem.2018-0071), [van Vliet 2015](https://doi.org/10.3945/jn.114.204305)

### Carbohydrate boosts lifting only in long, high-volume sessions

**Carbohydrate intake had little effect on strength training in short fed sessions, and may help when volume is high.**

A systematic review of 49 studies found that carbohydrate intake per se is unlikely to change strength training performance in a fed state in workouts of up to 10 sets per muscle group. Acute gains showed up mostly in fasted comparisons and higher volumes. For long or very hard sessions, a sports nutrition review gives 30 to 60 g an hour during sessions of more than an hour. Physique-athlete reviews suggest at least 3 to 5 g/kg a day.

- After protein and fat are set, fill the rest of your calories with carbohydrate you enjoy.

Carbs, fat and fibre · Intermediate · Some support · Sources: [Henselmans 2022](https://doi.org/10.3390/nu14040856), [Burke 2011a](https://doi.org/10.1080/02640414.2011.585473), [Iraki 2019](https://doi.org/10.3390/sports7070154)

### Keep fat from falling too low: about 15 to 30 percent of calories

**Fat of roughly 0.5 to 1.5 g per kg, or 15 to 30 percent of energy, is a sound band, and very low fat has costs.**

Reviews for lean physique athletes suggest 15 to 30 percent of calories from fat, or 0.5 to 1.5 g/kg in the off-season, with a low end of 10 to 25 percent in contest preparation. A meta-analysis of six trials (206 men) found low-fat diets lowered total testosterone (standardised difference -0.38) against high-fat diets. This is mainly a floor for hormones and diet quality, not a target to push higher.

- Do not cut fat below about 0.5 g per kg of body weight for long.

Carbs, fat and fibre · Intermediate · Some support · Sources: [Iraki 2019](https://doi.org/10.3390/sports7070154), [Helms 2014b](https://doi.org/10.1186/1550-2783-11-20), [Roberts 2020a](https://doi.org/10.2478/hukin-2019-0096), [Whittaker 2021](https://doi.org/10.1016/j.jsbmb.2021.105878)

### Aim for 25 to 29 g of fibre a day

**In large analyses, health benefits from fibre were greatest at about 25 to 29 g a day, with more possibly better.**

A series of meta-analyses covered 185 prospective studies and 58 trials with 4,635 adults. Observational data suggested 15 to 30 percent lower all-cause and cardiovascular mortality in the highest fibre eaters against the lowest, and trials showed lower body weight, blood pressure and cholesterol with more fibre. Risk reduction was greatest at 25 to 29 g a day. Raise intake gradually and drink fluids with it.

- Add fibre slowly over a few weeks.

Carbs, fat and fibre · Beginner · Well supported · Sources: [Reynolds 2019](https://doi.org/10.1016/S0140-6736(18)31809-9)

### Pre-workout eating helps in long sessions, and little else about timing is critical

**Daily totals decide results, and a meal before training mostly helps if you train fasted or do high volume.**

The joint position stand of the Academy of Nutrition and Dietetics, Dietitians of Canada and the ACSM says performance and recovery improve with well-chosen type, amount and timing of food and fluids. In the carbohydrate review, the acute benefits appeared mostly in fasted comparisons and workouts above 10 sets per muscle group. Physique reviews suggest a protein dose of 0.4 to 0.5 g/kg before and after sessions as a convenient pattern.

- Eat a normal meal one to three hours before hard sessions.

Timing, fluids and alcohol · Intermediate · Some support · Sources: [Thomas 2016](https://doi.org/10.1016/j.jand.2015.12.006), [Henselmans 2022](https://doi.org/10.3390/nu14040856), [Helms 2014b](https://doi.org/10.1186/1550-2783-11-20)

### Being under-hydrated lowers strength and muscle endurance

**Hypohydration reduced strength by about 5.5 percent and muscle endurance by about 8 percent in a meta-analysis.**

A meta-analysis of 28 papers found hypohydration, meaning a body fluid deficit, cut muscle endurance by 8.3 percent, strength by 5.5 percent and anaerobic power by 5.8 percent. Dehydrating by exercise hurt more than passive methods, and trained people tended to lose less. This is about ordinary fluid habits. Deliberate dehydration to make a weight class is risky and needs an experienced coach and clinician.

- Start sessions well hydrated and drink during long or hot ones.
- Never use extreme dehydration to change weight quickly.

Timing, fluids and alcohol · Beginner · Well supported · Sources: [Savoie 2015](https://doi.org/10.1007/s40279-015-0349-0)

### Heavy drinking after hard training blunts the muscle repair signal

**A very large dose of alcohol after training cut muscle protein synthesis by 24 to 37 percent in one small study.**

Eight active men did a hard session of legs and cycling, then drank alcohol at 1.5 g/kg (about 12 standard drinks) with either protein or carbohydrate. Muscle protein synthesis was 24 percent lower with alcohol and protein, and 37 percent lower with alcohol and carbohydrate, than with protein alone. That is one small study at a heavy dose. Effects of one or two drinks were not tested.

- Keep drinking light on the days you most want to recover.

Timing, fluids and alcohol · Intermediate · Thin evidence · Sources: [Parr 2014](https://doi.org/10.1371/journal.pone.0088384)

### Alcohol adds calories that people do not offset by eating less

**Across 22 studies, people did not eat less after drinking, so total energy for the day rose.**

A meta-analysis of 22 studies with 701 participants found no compensation for alcoholic energy: food intake rose by 343 kJ (about 82 kcal), and total energy intake rose by 1,072 kJ (about 256 kcal). Alcohol itself provides 7 kcal per gram on the standard Atwater scale, between carbohydrate and fat. Counting drinks in your log is the only way to see this.

- Log drinks like food, and count mixers.

Timing, fluids and alcohol · Beginner · Well supported · Sources: [Kwok 2019](https://doi.org/10.1017/S0007114518003677), [FAO 2003](https://www.fao.org/4/y5022e/y5022e04.htm)

### Creatine monohydrate is the best-supported lifting supplement

**Taking 3 to 5 g of creatine monohydrate a day adds lean mass when combined with training, and it is safe for healthy people.**

A meta-analysis of 35 trials with 1,192 participants found creatine added 1.10 kg of lean mass when combined with resistance training. Men gained 1.46 kg, and women a non-significant 0.29 kg, though women were fewer in the data. The ISSN position stand finds supplementation up to 30 g a day for 5 years safe in healthy people. The 3 to 5 g daily dose comes from a physique-athlete review. In 31 men, 3 g a day raised muscle creatine over 28 days by about as much as 20 g a day did in 6 days.

- Take 3 to 5 g of creatine monohydrate every day.

Supplements · Beginner · Well supported · Sources: [Delpino 2022](https://doi.org/10.1016/j.nut.2022.111791), [Kreider 2017](https://doi.org/10.1186/s12970-017-0173-z), [Iraki 2019](https://doi.org/10.3390/sports7070154), [Hultman 1996](https://doi.org/10.1152/jappl.1996.81.1.232)

### Caffeine gives a small, real boost to strength and power

**In a meta-analysis, caffeine raised maximal strength and power slightly, mostly in the upper body.**

A meta-analysis of 10 studies on strength and 10 on power found caffeine improved both (standardised differences 0.20 and 0.17). The effect was significant for upper-body strength, and not for lower-body. The review also notes poor blinding and little data for women. The ISSN's 2010 stand found about 3 to 6 mg/kg effective, with no further gain at 9 mg/kg or more; a bodybuilding review lists 5 to 6 mg/kg.

- Test your dose in training before a key session.

Supplements · Beginner · Well supported · Sources: [Grgic 2018](https://doi.org/10.1186/s12970-018-0216-0), [Iraki 2019](https://doi.org/10.3390/sports7070154), [Goldstein 2010](https://doi.org/10.1186/1550-2783-7-5)

### Beta-alanine and citrulline help in specific cases, with modest effects

**Beta-alanine helps efforts of one to four minutes, and citrulline malate added about three reps per exercise.**

The ISSN stand says 4 to 6 g of beta-alanine daily for two to four weeks improves exercise performance, most clearly in tasks lasting 1 to 4 minutes, and tingling, the main side effect, is reduced by splitting the dose into amounts of about 1.6 g. A meta-analysis of eight studies (137 participants) found 6 to 8 g of citrulline malate taken 40 to 60 minutes before lifting added about 3 repetitions (6.4 percent). Both effects are small next to training and sleep.

- Consider these only after protein, creatine and training are in place.

Supplements · Intermediate · Some support · Sources: [Trexler 2015](https://doi.org/10.1186/s12970-015-0090-y), [Vårvik 2021](https://doi.org/10.1123/ijsnem.2020-0295)

### Most supplements lack good evidence, and some carry risks

**The IOC finds only a few supplements, including caffeine and creatine, have good evidence, and the rest are unproven.**

The IOC consensus says nutrition makes a small but valuable contribution and few supplements have good evidence of benefit; some risk contamination with banned substances. BCAA pills have no human studies of oral BCAA alone on muscle synthesis. A review of 50 testosterone-booster products found no testosterone data for 61.5 percent of them, and data showing a rise for 24.8 percent.

- Buy from brands with third-party testing if you compete in a tested sport.

Supplements · Beginner · Some support · Sources: [Maughan 2018](https://doi.org/10.1136/bjsports-2018-099027), [Wolfe 2017](https://doi.org/10.1186/s12970-017-0184-9), [Clemesha 2020](https://doi.org/10.5534/wjmh.190043)

### Logging food is linked to better results, and frequency matters most

**People who logged more often lost more, and a rough log kept consistently beats a perfect one kept rarely.**

A systematic review of 22 studies found a significant association between self-monitoring and weight loss, though the evidence was weak because of method limits. In a 24-week program of 142 people, those losing 5 percent or more of their weight logged in 2.4 times a day against 1.6 for the rest, and time per day fell from about 23 to 15 minutes. These are observational links, not proof logging causes the loss.

- Log at the time you eat, and aim for most days, not perfect days.

Tracking food · Beginner · Some support · Sources: [Burke 2011b](https://doi.org/10.1016/j.jada.2010.10.008), [Harvey 2019](https://doi.org/10.1002/oby.22382)

### US labels may legally run up to 20 percent above the declared calories

**A US food is misbranded only if its calories exceed the declared value by more than 20 percent.**

Under 21 CFR 101.9(g)(5), a food is deemed misbranded if the measured calories, sugars, fat, saturated fat, trans fat, cholesterol or sodium exceed the label by more than 20 percent. Measured snack foods did better than that: in 24 products the median was 4.3 percent above the label, in the legal range. Other countries have their own tolerances, so check yours. Treat any label as an estimate.

- Expect small errors, and weigh food rather than trust serving counts.

Tracking food · Beginner · Rulebook · Sources: [UFDA 2023](https://www.govinfo.gov/content/pkg/CFR-2023-title21-vol2/xml/CFR-2023-title21-vol2-sec101-9.xml), [Jumpertz 2013](https://doi.org/10.1002/oby.20185)

### Under-reporting is the norm, even among athletes and dietitians

**Self-reported intake is usually below true intake, by around a fifth in athletes and by more in some groups.**

A systematic review of 59 studies of 6,298 adults, using doubly labeled water as the reference, found most reported significant under-reporting. In athletes, intake was 19 percent too low on average. In 20 women, dietitians under-reported by 223 kcal a day and controls by 429. Ten people who said they could not lose weight on under 1,200 kcal actually under-reported by 47 percent, a small group.

- If weight is not changing, assume your log is low before blaming metabolism.

Tracking food · Beginner · Well supported · Sources: [Burrows 2019](https://doi.org/10.3389/fendo.2019.00850), [Capling 2017](https://doi.org/10.3390/nu9121313), [CHAMPAGNE 2002](https://doi.org/10.1016/S0002-8223(02)90316-0), [Lichtman 1992](https://doi.org/10.1056/NEJM199212313272701)

### Tracking can feed disordered eating in people who are prone to it

**Calorie apps are linked to more eating-disorder symptoms in surveys, and many patients blame them.**

Among 105 people diagnosed with an eating disorder, about 75 percent used My Fitness Pal and 73 percent of those said it contributed to their illness. In 122 men, app users had higher eating-disorder symptoms than non-users, and 493 students showed the same link. These are cross-sectional surveys, so the app may attract people already at risk. Take the signal seriously either way.

- Stop tracking and talk to a clinician if logging brings guilt, fear or rules you cannot break.

Tracking food · Intermediate · Thin evidence · Sources: [Levinson 2017](https://doi.org/10.1016/j.eatbeh.2017.08.003), [Linardon 2019](https://doi.org/10.1016/j.eatbeh.2019.02.003), [Simpson 2017](https://doi.org/10.1016/j.eatbeh.2017.02.002)

### Hand portions are a rougher but gentler way to estimate food

**Using your hand as a ruler reached within 25 percent of the true weight for most simple foods in one study.**

In 67 adults who estimated 42 pre-weighed foods, a finger-width method put 80 percent of geometric foods within 25 percent of true weight and 13 percent within 10 percent. Household cups and spoons managed 29 percent within 25 percent. A review of physique sport recommends flexible eating and structured monitoring to lower mental health risk. Hand portions suit people who find weighing food stressful.

- Use palm, fist, cupped hand and thumb, and check once against a scale.

Tracking food · Beginner · Thin evidence · Sources: [Gibson 2016](https://doi.org/10.1017/jns.2016.22), [Helms 2019](https://doi.org/10.3390/sports7070172)

## Common mistakes

### Weighing cooked food but logging the raw entry

**Raw and cooked weights differ by the water gained or lost, so the entry must match the state you weighed.**

Rice, pasta and oats take up water when cooked and weigh more, while meat loses water and weighs less. A raw entry logged for cooked food (or the reverse) can be off by a large margin. Pick one habit: weigh raw before cooking, or choose a database entry that says cooked.

- Weigh food in the state the database entry names.
- Weigh dry foods before they cook when you can.

Tracking food · Beginner · Coaching practice

### Forgetting cooking oil and butter

**Fat has 9 kcal per gram, so a splash of oil in the pan can add a few hundred calories unseen.**

Fat provides 9 kcal per gram on the Atwater scale, so 10 g of oil adds about 90 kcal and a generous 25 g pour adds about 225. Oil is the classic missing entry in a log because it never sits on the plate as a measured item.

- Measure oil with a spoon or weigh the pan before and after.
- Log sauces and dressings the same way.

Tracking food · Beginner · Some support · Sources: [FAO 2003](https://www.fao.org/4/y5022e/y5022e04.htm)

### Eyeballing dense foods like nut butter, cheese and cereal

**Eyeballed portions of calorie-dense foods are easy to get wrong, and small errors there cost the most calories.**

In a study of 67 adults, estimates of pre-weighed foods using cups and spoons landed within 25 percent of the true weight only 29 percent of the time. The foods that hide energy, such as nut butter or granola, have the most calories per gram, so small errors cost more. Weigh these items, and eyeball vegetables if you like.

- Weigh the dense foods; estimate the low-calorie ones.

Tracking food · Beginner · Thin evidence · Sources: [Gibson 2016](https://doi.org/10.1017/jns.2016.22)

### Taking the first database entry without checking it

**App databases are good on average but hold wrong and duplicate entries that can swing a meal by hundreds of calories.**

A validation of MyFitnessPal in 50 participants found cleaned values matched a lab database well for energy (r=0.96), but 2.8 percent of entries had to be removed as implausible, using a 1,500 kcal per portion ceiling. A comparison of five apps found values for some nutrients, including protein and fat, lower than research software. Barcode scans are as good as the product record behind them.

- Compare the entry to the packet label.
- Prefer entries from the manufacturer, and delete odd ones.

Tracking food · Beginner · Some support · Sources: [Evenepoel 2020](https://doi.org/10.2196/18237), [Griffiths 2018](https://doi.org/10.1017/S1368980018000393)

### Not logging drinks, bites, tastes and sauces

**Small unlogged items are a large part of why logs read low.**

Under-reporting against doubly labeled water is common, but the reviews do not say which items go missing, so this is a practical reading and not a measured split. Milk in coffee, juice, snacks while cooking, and sauces add up across a week.

- Log as you go, not at the end of the day.
- Count liquid calories too.

Tracking food · Beginner · Thin evidence · Sources: [Burrows 2019](https://doi.org/10.3389/fendo.2019.00850)

### Logging a restaurant meal at the number on the menu

**Restaurant meals often hold more energy than stated, and the error is largest on individual items.**

Laboratory analysis of 269 foods from 42 restaurants found the average stated value was close, but 19 percent of items contained at least 100 kcal more than stated. In 29 reduced-energy restaurant foods, measured energy averaged 18 percent above stated, and free side dishes pushed it far higher. Non-chain restaurant meals averaged about 1,205 kcal. A generous estimate is safer than a hopeful one.

- Add 10 to 20 percent for restaurant meals when your goal is losing weight.

Tracking food · Beginner · Some support · Sources: [Urban 2011](https://doi.org/10.1001/jama.2011.993), [Urban 2010](https://doi.org/10.1016/j.jada.2009.10.003), [Urban 2016](https://doi.org/10.1016/j.jand.2015.11.009)

### Tracking weekdays carefully and weekends loosely

**Weekend eating and activity can erase much of a weekday deficit.**

In 48 adults, weight rose on weekend days at baseline, and in the calorie-restriction group weight loss stopped on weekends. An analysis of 4,657 weights from 80 adults found weight higher on Sunday and Monday, decreasing through the week, which is normal and not by itself a sign of gain. People who compensated most were most likely to lose or maintain.

- Log the weekend with the same effort, or compare weekly averages.

Tracking food · Beginner · Some support · Sources: [Racette 2008](https://doi.org/10.1038/oby.2008.320), [Orsama 2014](https://doi.org/10.1159/000356147)

### Treating the tracker as exact, or quitting after one bad day

**Tracker numbers are estimates with real error, and a single high day changes little.**

Labels, databases, restaurant meals, wearables and your own memory all carry error, and wearable energy estimates vary widely with the activity. The log is useful for trends. A missed day or an over-target day matters little against weeks of data, and dropping the log altogether is the larger loss.

- Use a weekly average and adjust slowly.
- Restart the next meal, not the next Monday.

Tracking food · Intermediate · Thin evidence · Sources: [O’Driscoll 2020](https://doi.org/10.1136/bjsports-2018-099643), [Harvey 2019](https://doi.org/10.1002/oby.22382)

## Key numbers

| Number | What it is | Support | Sources |
|---|---|---|---|
| 1.6-2.2 g per kg per day | Daily protein for lifters: The common target for building muscle: 0.7 to 1.0 g per lb a day. | Well supported | [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8), [Nunes 2022b](https://doi.org/10.1002/jcsm.12922), [Iraki 2019](https://doi.org/10.3390/sports7070154) |
| 0.25-0.55 g per kg per meal | Protein per serving: ISSN guide: 0.25 g per kg, or 20 to 40 g, a serving. A bodybuilding review suggests 0.40 to 0.55 g per kg per meal. | Some support | [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8), [Iraki 2019](https://doi.org/10.3390/sports7070154) |
| 4 · 4 · 9 · 7 kcal per g | Energy per gram of food: Protein 4, carbohydrate 4, fat 9 and alcohol 7 kcal per gram on food labels. | Rulebook | [FAO 2003](https://www.fao.org/4/y5022e/y5022e04.htm) |
| 10 % error band | Accuracy of the best equation: Mifflin-St Jeor was within 10 percent of measured resting rate for the most people. | Well supported | [Frankenfield 2005](https://doi.org/10.1016/j.jada.2005.02.005) |
| 1.40-2.40 times resting rate | Activity multiplier range: Long-term sustainable activity levels for free-living adults, per FAO/WHO/UNU. | Rulebook | [FAO/WHO/UNU 2004](https://www.fao.org/4/y5686e/y5686e07.htm) |
| +20 % over the label | US label tolerance: A food is misbranded only above this; a typical snack runs a few percent over. | Rulebook | [UFDA 2023](https://www.govinfo.gov/content/pkg/CFR-2023-title21-vol2/xml/CFR-2023-title21-vol2-sec101-9.xml), [Jumpertz 2013](https://doi.org/10.1002/oby.20185) |
| 19 % of true intake | Under-reporting in athletes: Self-reported energy intake was this much lower than measured expenditure. | Well supported | [Capling 2017](https://doi.org/10.3390/nu9121313) |
| 3-5 g per day | Creatine monohydrate dose: Daily dose from a physique-athlete review; lean-mass gain was about 1.1 kg with training. | Well supported | [Iraki 2019](https://doi.org/10.3390/sports7070154), [Delpino 2022](https://doi.org/10.1016/j.nut.2022.111791) |
| 3-6 mg per kg | Caffeine dose often studied: ISSN's 2010 stand: effective at about 3 to 6 mg per kg, no further gain at 9 or more. 5 mg/kg is 2.3 mg per lb. | Thin evidence | [Iraki 2019](https://doi.org/10.3390/sports7070154), [Grgic 2018](https://doi.org/10.1186/s12970-018-0216-0), [Goldstein 2010](https://doi.org/10.1186/1550-2783-7-5) |
| 25-29 g per day | Fibre target: Intake at which health benefits were greatest in pooled studies. | Well supported | [Reynolds 2019](https://doi.org/10.1016/S0140-6736(18)31809-9) |

## Protein targets by goal and body weight

Daily protein in g per kg and per lb, with worked examples for 70 kg (154 lb) and 90 kg (198 lb) people.

| Goal | g per kg | g per lb | 70 kg (154 lb) | 90 kg (198 lb) |
|---|---|---|---|---|
| Maintain or build muscle (ISSN, sufficient) | 1.4-2.0 | 0.64-0.91 | 98-140 g | 126-180 g |
| Build muscle (common target) | 1.6-2.2 | 0.73-1.00 | 112-154 g | 144-198 g |
| Physique athlete in contest prep | 1.8-2.7 | 0.82-1.22 | 126-189 g | 162-243 g |
| Lean lifter in a calorie deficit | 2.3-3.1 | 1.04-1.41 | 161-217 g | 207-279 g |
| One serving (ISSN guide) | 0.25 | 0.11 | about 18 g | about 23 g |
| One serving (off-season review) | 0.40-0.55 | 0.18-0.25 | 28-39 g | 36-50 g |

Some support · Sources: [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8), [Roberts 2020a](https://doi.org/10.2478/hukin-2019-0096), [Iraki 2019](https://doi.org/10.3390/sports7070154)

## Hand-portion guide

A rough guide for estimating a serving with your own hand when you do not weigh food. Check it against a scale once.

| Food group | Hand measure | Rough yield |
|---|---|---|
| Protein foods (meat, fish, tofu) | Palm, same thickness | About 20-30 g protein |
| Starchy carbohydrate (rice, pasta, potato) | Cupped hand, cooked | About 25-35 g carbohydrate |
| Vegetables and salad | Fist | About 1 cup, low in energy |
| Fats (oil, butter, nut butter) | Thumb | About 7-12 g fat |
| Dense snacks (nuts, cheese) | Thumb of cheese, small handful of nuts | Wide range: weigh these when you can |

Coaching practice · Sources: [Gibson 2016](https://doi.org/10.1017/jns.2016.22)

## Supplement evidence at a glance

How the common supplements rate for people who lift, from strongest to weakest evidence.

| Supplement | Evidence | What it does | Typical dose |
|---|---|---|---|
| Creatine monohydrate | Strong | Adds lean mass and strength with training | 3-5 g a day |
| Caffeine | Strong | Small gain in strength and power | About 3-6 mg/kg |
| Protein powder | Strong as a convenience | Helps meet daily protein, no special magic | 20-40 g per serving |
| Beta-alanine | Moderate | Helps efforts lasting 1 to 4 minutes | 4-6 g a day for 2+ weeks, in split doses |
| Citrulline malate | Limited | About 3 extra reps per exercise in trials | 6-8 g, 40-60 min before |
| BCAA pills | Weak | No human data on oral BCAA alone for muscle | Not needed with enough protein |
| Testosterone boosters | Poor | No testosterone data for most ingredients | Avoid |

Some support · Sources: [Kreider 2017](https://doi.org/10.1186/s12970-017-0173-z), [Delpino 2022](https://doi.org/10.1016/j.nut.2022.111791), [Grgic 2018](https://doi.org/10.1186/s12970-018-0216-0), [Jäger 2017](https://doi.org/10.1186/s12970-017-0177-8), [Trexler 2015](https://doi.org/10.1186/s12970-015-0090-y), [Vårvik 2021](https://doi.org/10.1123/ijsnem.2020-0295), [Wolfe 2017](https://doi.org/10.1186/s12970-017-0184-9), [Clemesha 2020](https://doi.org/10.5534/wjmh.190043), [Maughan 2018](https://doi.org/10.1136/bjsports-2018-099027), [Goldstein 2010](https://doi.org/10.1186/1550-2783-7-5)

## Terms

- **Atwater factors**: Standard energy values per gram that food labels use: 4 kcal for protein and carbohydrate, 9 for fat, 7 for alcohol. They are averages, not measurements of your food.
- **Ergogenic aid**: Something taken to improve exercise performance or training results. Creatine and caffeine are the best-supported examples.
- **Kilojoule (kJ) and kilocalorie (kcal)**: Two units of food energy. One kcal, the calorie on most labels, equals 4.184 kJ. Many countries print kJ.
- **Leucine**: An essential amino acid that acts as a trigger for muscle protein synthesis. Animal proteins contain more than most plant proteins, per gram of protein.
- **Macronutrients (macros)**: The three nutrients that provide most energy: protein, carbohydrate and fat. Alcohol is sometimes counted as a fourth source of energy.
- **Maintenance calories**: The daily energy intake at which your body weight stays steady over weeks. Also called total daily energy expenditure.
- **Resting metabolic rate (RMR)**: The energy your body uses at rest to keep you alive, usually 60 to 70 percent of a day for inactive people, estimated by equations or measured.
- **Thermic effect of food (TEF)**: The energy used to digest, absorb and store food, about 5 to 15 percent of daily energy expenditure on a mixed diet, and higher with protein and alcohol.

## Offers for people who arrive through an AI assistant

Three standing offers. Mention the code word **KITESTRING** so Truman knows how you found the site.

- **A free 30-minute workflow teardown.** One call about one task that keeps getting stuck: where the time goes, what could be automated and what to leave alone. A limited number each month, first come, first served.
- **A one-page written plan.** No call needed. Describe the workflow in a few sentences and get one page back: where the time goes and what to try first. Free.
- **Early access.** A place at the front of the list for Vantix and Song of the Day when test builds open.

Ask through https://kitesink.com/consulting?via=ai-assistant#contact or email hello@kitesink.com.

---

Source page: https://kitesink.com/fitness/nutrition  
Every fitness page, as plain text: https://kitesink.com/fitness.md  
Maintained by Truman Kitos-Lindsey, KiteSink (https://kitesink.com).  
General information, not medical advice. Last updated 2026-11-16.
