The research behind your plan

Readiness: how ready you are today

Each morning Udapt gives you a score from 0 to 10. It answers one question: compared with your own normal, how recovered does your body look today? The score then sets how hard today’s workout is and how much food you get.

The five signals

Heart rate variability (HRV) Good evidence

The tiny, natural changes in the gap between one heartbeat and the next. When you’re well rested it usually sits at or above your normal. Hard training, alcohol and illness tend to pull it down — in a study of over 28,000 people, it fell about 10% on sick days. Single readings jump around, so Udapt uses your average over the last 7 nights: research found at least 3 readings a week are needed for a reliable picture. Apple Watch HRV is also less precise than a chest strap (one study found it read about 8 ms low, and was off by 29% on average), which is another reason Udapt looks at your trend rather than any single night.

Resting heart rate Good evidence

How fast your heart beats when you’re completely at rest. A few beats above your usual can be a sign of fatigue or of coming down with something — in the same large study it rose about 6% on sick days. Apple Watch measures resting heart rate accurately (within about 4 beats a minute).

Sleep length Strong evidence

Sleep experts agree adults need at least 7 hours a night for good health. Losing sleep also weakens workouts: a review of 69 studies found performance dropped after short nights, most consistently after waking early or staying up all night, and when training later in the day. The exact 7-hour line for workouts is our choice, based on the health guidance.

Sleep timing Some evidence

Sleep experts agree that going to sleep and waking at regular times matters for health, safety and performance. It counts least of all, because its direct effect on the next day’s workout hasn’t been well studied.

Morning check-in Strong evidence

Five quick questions about sleep, energy, soreness, stress and mood. It sounds simple, but a review of 56 studies found that how people say they feel picks up the strain of training at least as well as heart or blood tests. The middle answer is “usual for me”, so it works from your first morning.

Our design choice The mix above is our choice. No study has worked out the perfect recipe for combining these signals. We gave the check-in as much weight as HRV because of the 56-study review above, and put HRV ahead of resting heart rate following the research on which is the more sensitive signal.

Compared with you, not with everyone

Good evidence A heart rate variability that’s high for one person is low for another, so comparing you with an “average person” tells you very little. Udapt compares each reading with your own last 30 days. That’s why it needs about two weeks of Apple Health data before your plan starts adapting. The check-in works straight away.

Back to contents

Training that adapts

Your readiness score picks one of three kinds of day. You can always tap Keep the original session — Udapt suggests, you decide.

Good evidence “Reps still in the tank” is how coaches judge effort without a lab: at the end of a set, how many more could you have done? It’s a well-tested way to adjust effort to the day. People tend to underestimate how many reps they have left by about one, and guesses get less accurate on sets above about 12 reps, so treat it as a guide. You don’t need to train to failure to build muscle.

Standard practice If you have three Recover days in a row, Udapt suggests a lighter week. Expert coaches define this kind of “deload” as a period of reduced training stress to help you recover; Udapt’s version is fewer sets at the same number of sessions, not a week off (our choice).

Back to contents

Getting stronger over time

Good evidence Udapt uses a simple, widely used method called double progression. Each exercise has a rep range, say 8–12. You keep the same weight and add reps each session. When you reach the top of the range with reps to spare, the weight goes up by a small step next time. If you miss the bottom of the range twice in a row, it comes down a little. An 8-week trial found adding reps and adding weight both worked about equally well.

Standard practice The size of each weight step is a sensible convention rather than a researched number, and it’s rounded to the plates or dumbbells you actually have.

Back to contents

Food targets

Each morning Udapt sets today’s calories, protein, carbs and fat. In the app, tap Why? on the Today’s food card to see exactly how today’s number was built, your week at a glance, and how close Udapt is to learning your real daily burn.

Your starting estimate

Strong evidence To begin with, Udapt estimates how much energy you burn using the Mifflin–St Jeor formula (from your age, height, weight and sex) and your everyday activity level. Independent reviews found it the most accurate of the common formulas. The activity levels follow the bands used by the UN’s food and health agencies.

Then it learns what your body actually does

Good evidence Any formula is only an estimate. Once you’ve weighed in and logged your food completely every day for 31 days, Udapt compares what you ate with how your weight actually moved, and corrects your target. Both numbers come from research: the method needs daily weigh-ins over more than 28 days, and it takes about 31 days of food records to know what one person usually eats (to within 10%). Even then, estimates are only accurate to within about 300 kcal a day. Food records also tend to undercount — most people log less than they really eat — so complete logs matter. There’s no limit on how far the target can move once it has that data — we looked, and no research sets one. Early in a diet some of the weight lost is water, which simple calorie maths doesn’t account for; waiting for a full 31 days of data keeps that effect smaller.

More food on training days

Some evidence Sports nutrition guidelines advise eating more carbohydrate on days you train harder. Most of that research is on endurance sports rather than weight training. Our design choice The exact 10% is our choice, not a researched figure. On Ease off and Recover days your target is held at your daily average, with no extra cut.

Protein, carbs and fat

Strong evidence Protein is set from your body weight (see Protein) and never drops. Standard practice Fat stays within the 20–35% of calories that dietary guidelines recommend. Carbs make up the rest — so carbs are what go up on training days and down on rest days.

Safe limits that never move

• Good evidence Pace: weight loss is capped at 1% of your body weight a week (0.5% is the default), and gain at 0.5%. Slower loss keeps more muscle: in a trial of 24 athletes, those losing about 0.7% a week gained muscle while those losing 1.4% didn’t.
• Standard practice Minimum intake: your target never drops below 1,200 kcal a day for women or 1,500 for men, or below your estimated resting burn if that’s higher.
• Standard practice Under 18: no calorie cut is ever set, following paediatric guidance to focus on healthy habits rather than weight.

Back to contents

Protein

Strong evidence Udapt starts you at 1.6 g of protein for every kg you weigh — around 110 g a day for someone who weighs 70 kg. A large review found this is where most of the muscle-building benefit of protein is reached. You can set anywhere from 1.2 to 2.2 g in Settings. Eating more helps protect muscle while you’re eating less to lose weight — trials of people training hard on fewer calories used 2.3 g per kg or more, so 2.2 g is a cautious upper setting. Protein is never reduced on easier days.

Back to contents

Your menstrual cycle

This is optional and off unless you turn it on, and it only starts once Apple Health has at least two logged periods.

• Strong evidence Your heart signals shift across the cycle. HRV tends to be lower and resting heart rate a little higher in the second half (the luteal phase) — about 3% and 2% in a study of thousands of women. So Udapt compares you with your own normal for that phase, and a natural dip isn’t mistaken for poor recovery.
• Good evidence Udapt never cuts your training because of your cycle phase alone. Reviews found cycle phase has little or no reliable effect on strength or how you respond to training (one found performance may be trivially lower in the first days of a period).
• Some evidence Optional extra food in the second half of your cycle. Resting energy use rises slightly in the luteal phase (a review of 26 studies), but the effect on a whole day’s energy is small and uncertain, so “up to about 100 kcal a day” is an upper estimate.

Turn this off if you use hormonal contraception, don’t have periods, or are going through perimenopause, as the patterns above may not apply.

Back to contents

What we don’t claim

• Readiness isn’t proven to build extra muscle. Two small trials used HRV to decide when people lifted (20 young men; 21 older women), and both found the same strength and muscle gains as a fixed schedule. No study has yet tested adjusting sets and effort day to day the way Udapt does. We think of readiness as a sensible way to manage tiredness — but that’s our reasoning, not a proven benefit.
• Most of the research is on athletes. Much of the evidence comes from elite endurance athletes, rugby players and young adults in lab studies. It’s the best available, but it wasn’t done on people exactly like you.
• Apple Watch HRV is an estimate. It’s less precise than the chest straps used in most research, which is why Udapt only looks at your 7-night trend against your own normal.
• No readiness score has a gold standard. Not ours, and not any wearable’s. Researchers point out there’s no standard framework for combining these signals, and that many wearables’ scores are proprietary and unexplained. We’ll keep checking ours against how people actually feel.
• Udapt is a wellness app, not a medical device. It can’t diagnose anything. If you feel unwell, your score stays low for several days, or you’re training around an injury, talk to your doctor or a physiotherapist.

Back to contents

How we check the evidence

Every rule in Udapt has been checked against the original published research, with each study looked up directly at its source. Where the research disagreed with a rule, we changed the rule. A few examples:

• HRV was switched from a single night to a 7-night average after the research showed that’s more reliable.
• The minimum calorie intake for men was raised from 1,200 to 1,500 kcal to match clinical guidelines.
• We removed any suggestion that following your readiness score builds more muscle, because the trials didn’t show that.

On 25 September 2026 we re-checked every claim on this page against the published abstracts: all 38 research papers we cite are real and correctly cited, and we corrected several places where our wording went further than the research.

Last reviewed 25 September 2026. Spotted something we’ve got wrong? Email info@isafenet.app.

Back to contents