Pensum
How accurate is a TDEE calculator?
A calculator turns four facts about your body into a maintenance number. A few weeks of your own weigh-ins and food diary show how close it was.
A TDEE calculator gives a reasonable first guess at your maintenance calories. The error in that guess comes from two places, the equation for your resting metabolism and the activity level you pick, and for one person the two together can add up to a couple of hundred calories a day. The scale won't show which way you're off for a few weeks.
What a TDEE calculator estimates
TDEE, total daily energy expenditure, is the calories you use in a full day. It adds up your resting metabolism (BMR), the energy spent digesting food, and everything you do on your feet, from workouts to walking to the kitchen. A calculator gets there in two steps. It predicts BMR from height, weight, age and sex, then multiplies that by an activity factor you pick from a list.
Many calculators run Mifflin-St Jeor, an equation from 1990. The Pensum TDEE calculator uses a newer regression equation whose weight term grows more slowly than body weight, so the estimate doesn't run away for heavier people the way a flat 10 kcal per kilogram does. If you enter your body fat, it switches to a second equation from the same series, built on fat-free mass and fat mass. Either way, the result is a prediction for an average person with your numbers.
How far off a TDEE calculator can be for one person
A 2005 systematic review compared the common equations for resting metabolism with measured values in healthy adults. Mifflin-St Jeor came out as the most reliable of them, and it still produced noteworthy errors for individuals. That review covered BMR alone. A TDEE calculator puts the activity guess on top of it.
With both guesses combined, two people with identical stats can burn a couple of hundred calories a day apart while every formula gives them the same number. That's enough to decide whether a cut works. Losing half a kilogram a week takes about 550 kcal a day under maintenance. If your real maintenance sits a couple of hundred below the calculated one, a large part of the planned deficit is gone before you start.
Body fat can make the guess more personal if the figure is good. A DEXA scan or a skilled caliper measurement is close enough to help. Smart scales and handheld devices measure electrical resistance, which shifts with hydration, and they're often several percentage points off. A wrong figure moves both BMR and protein, so leave the field empty unless you trust the number.
Why activity multipliers are the weak link
The activity factor multiplies everything the BMR equation produced. The usual list runs from 1.2 for a desk job with no exercise through 1.375, 1.55 and 1.725 up to 1.9 for hard training twice a day or a physical job. Each step up the list adds a fixed share of your BMR to every day of the estimate.
The levels count training days. None is sedentary, one to three is light, three to five is moderate. Two people who both train four days a week can spend very different amounts of energy in those sessions and in how much they walk the rest of the day, and the factor treats them the same. Most people overrate their activity, so when you're between two levels, take the lower one.
Why a single weigh-in means little
After a week on the calculated target it's tempting to judge the number by the scale. One weigh-in can't tell you whether it's right. The scale can move a kilogram between two mornings with no change in fat, and most of that is water. Glycogen holds roughly three to four grams of water per gram stored, so a weekend of higher carbs can add one to three kilograms that leave again a few days later.
The menstrual cycle adds its own swing. Around a period the scale usually rises by about half a kilogram, all of it water, highest on the first day of bleeding and gone within a week. A tracker that adjusts your target from your weight reads that as stored energy unless it knows the dates. Period tracking in Pensum explains how that water is kept out of the trend.
The expenditure page has a chart of sixty days at a flat weight. The raw daily weigh-ins swing across more than 2 kg on water alone, while the smoothed trend stays inside about half a kilogram. Pensum's trend moves 10% of the way toward each new weigh-in, which gives it a half-life of about a week. A salty dinner barely moves it, and a real trend shows within days.
What to do after the first weeks of data
Stay on the calculated target for now, and keep logging and weighing in the way you normally would. Pensum then replaces the formula with an adaptive TDEE: it compares what you ate with how your weight trend moved over the last two weeks and works out what you actually burn.
With less than a full week of data the app shows the formula number untouched. For the first month the estimate runs faster, so a wrong starting point gets corrected quickly. In a simulated test where the formula started 15% wrong, the estimate ended about 3 kcal from the true value. Within a few weeks the measured figure takes over, and your calorie target follows it.
Log consistently. Self-reported intake runs around 20% light on average in validation studies, but your target is computed in the same units, so a steady bias leaves the deficit where it should be. A change in how you log, such as moving from photo estimates to a kitchen scale, shifts the number for a couple of months. Give it time as well. A real change in expenditure takes six to nine weeks to show halfway in the estimate, which is the cost of ignoring water swings.
After a month, check the trend against the number. At your logged intake, the measured expenditure implies how far your weight trend should move over the coming weeks. The usual conversion of 7,700 kcal per kilogram overstates long-term loss, because expenditure falls as you get lighter, and that's why the maintenance number needs re-measuring as you go.
From a calculator to a measured number
The TDEE calculator runs the same equations, activity factors and macro rules the Pensum app uses for your first targets. In the app that number only covers the first days. As your weigh-ins and diary build up, the measured expenditure replaces it, and the measurement is part of the free app, with no ads and no account. If you already have months of history in another tracker, importing it gives the estimate real data from the first day.