7-day balance: the calorie goal that doesn't reset on Monday
D-Fit now tracks the rolling average of your last 7 days instead of the calendar week. The science behind it, how the app works out your room for today, and why your body responds to the average, not to the day.
Your body doesn’t know what day of the week it is. It doesn’t close the books on Sunday night and it doesn’t open a fresh account on Monday morning. What it feels is the average of what comes in over time, and that average is what your weight responds to.
That idea is what gave D-Fit a new feature: the 7-day Balance. Instead of adding up calories from Monday to Sunday, the app now looks at a rolling window of your last 7 completed days, works out the average and uses it to tell you how much room you have today. A heavy Saturday still counts on Monday. Seven days later, it drops out on its own. Nothing resets, nothing piles up forever.
This article explains why a rolling average is the most faithful way to track a diet, what the research says about it and how the app does the math.
7-day Balance
One line to keep in place
Each bar is the difference between what you ate and that day's goal. The dashed line is the average of the last 7 completed days. The game is keeping it near zero.
Original goal: 2,000 kcal (−64 kcal)
Last 7 completed days
The heavy Saturday still counts on Monday. Seven days later it drops out on its own. Nothing resets, nothing piles up forever.
Quick answer
D-Fit tracks the rolling calorie average of your last 7 days. Each day is compared against that day’s goal, and the average of those differences becomes the line you want to keep near zero. From it, the app works out your room for today: the original goal, plus or minus a fraction of the window’s balance, divided by 7 and capped at ±300 kcal.
In practice, you only have one job: keep the line balanced. A day above, a few days slightly below. A day below, a little more on the days that follow. No countdown to Sunday, no starting from scratch on Monday.
Why the calendar week is an artificial unit
Most diet apps, and most of us in our heads, organise eating into weeks. It’s a useful convention for work and for the calendar, but it has nothing to do with physiology. Two problems show up when the calorie math follows the calendar:
1. Monday has one day of data. Any average calculated on Monday is the average of a single day. On Tuesday, of two. The information only becomes reliable around midweek, and by then the cycle is nearly over.
2. Sunday becomes the day to fix everything. If the week closed with a positive balance, the only way to settle the account is to eat far less on the last day. That concentrates all the pressure into one day, which is the opposite of what physiology asks for.
With a rolling window, both problems disappear by construction. Every day is midweek. The average always covers seven complete days, and the adjustment always spreads across the next seven.
What the science says about averages and timeframes
There are four lines of evidence pointing at the same conclusion: what matters is the cumulative balance over time, not the shape of any individual day.
1. Weight follows the cumulative balance, with a lag of months
The energy balance model most used in research, published by Kevin Hall and colleagues in The Lancet in 2011, shows that body weight responds to cumulative energy balance with a time constant of months, not days. A sustained deficit takes about a year to reach half of its final effect on weight, and how that deficit is distributed within a week barely changes the outcome.
Translation: to your body, eating 1,800 kcal every day or alternating 1,600 and 2,000 is the same thing. What it registers is the average.
2. Intermittent and continuous restriction give the same result when the average matches
The studies that compared continuous diets with intermittent diets are the most direct test of this idea. If the distribution within the week mattered, the results should diverge. They don’t.
Harvie et al., 2011 (International Journal of Obesity)
→ 107 women, 6 months
→ Continuous restriction (25% every day) vs 5:2 (two days of hard restriction)
→ Weight and fat loss: equivalent
Headland et al., 2016 (Nutrients), meta-analysis
→ Trials of 6 months or longer
→ Intermittent vs continuous: no significant difference in weight loss
Byrne et al., 2018 (International Journal of Obesity), MATADOR study
→ Blocks of 2 weeks in a deficit, 2 weeks at maintenance
→ More fat loss and LESS metabolic adaptation
than continuous restriction with the same cumulative deficitWhen the average matches, the result matches. MATADOR even suggests that alternating days and phases may be better than restricting equally every day, because it gives your metabolism moments of slack.
3. The weekend is where the average is decided
Two studies watched what happens to weight across the week in ordinary people. The pattern is consistent: intake rises from Friday to Sunday, weight rises with it, and it falls again from Monday to Thursday.
Racette and colleagues, in 2008, showed that weight gain over a year was concentrated on weekends. Orsama and colleagues, in 2014, followed daily weigh-ins and found the same weekly rhythm, with one important difference: the people who lost weight were the ones who compensated for the weekend during the week. Not the ones who had no weekend. The ones who balanced the average.
That is exactly what a rolling average measures. It doesn’t ask Saturday to look like Tuesday. It asks the seven days, together, to stay near the line.
4. What psychology knows about “starting over”
Here the evidence cuts both ways, and it’s worth being honest about both.
Polivy and Herman described, back in the 1980s, a pattern they called the “what-the-hell effect”: when someone on a diet goes past the limit they set for the day, the tendency is to eat even more afterwards, because the day is already “blown”. A rigid limit with a reset date creates exactly that trigger. If the account clears on Monday, any excess on Wednesday feels unrecoverable until then.
On the other side, Dai, Milkman and Riis showed in 2014 the fresh start effect: temporal landmarks like Mondays and the start of a month genuinely increase motivation to pursue goals. That’s real and useful.
D-Fit’s design choice was to separate the two. The math follows physiology: a rolling average, no reset. The motivation can follow the calendar: you can pick Monday to review your goal, try a new routine or weigh in more carefully, without that erasing what came before. The average neither punishes nor forgives. It just informs.
How D-Fit does the math
Let’s get to the mechanism, no black box.
The window
The window is the 7 complete days before today. If today is Wednesday, the window runs from last Wednesday to yesterday, Tuesday.
For each day in the window, the app calculates:
day deviation = calories eaten − that day's goalNote the “that day’s”. If your goal changed midway through the window, because adaptive TDEE suggested an adjustment or because you edited it by hand, each day is compared against the goal that was in force when it happened. Changing your goal today never rewrites yesterday’s result.
Days with no log stay out of the average: the app doesn’t assume you ate zero. When it works out your room for today, they count as a day on target, which is the most neutral assumption available. A day with no log never creates debt or credit.
The average
7-day average = sum of deviations on logged days ÷ logged daysThis is the headline number on the card and the dashed line on the graph. If it reads +64 kcal/day, you ate an average of 64 kcal above goal over the last seven days. Near zero is the target. It doesn’t have to be exactly zero: an average within ±50 kcal/day is, for all practical purposes, on the line.
Your room for today
Here’s the part that changes your day to day. The app takes the window’s total balance, divides it by 7 and applies it to today’s goal, with a cap:
today's adjustment = −(window balance ÷ 7), capped at ±300 kcal
room for today = original goal + today's adjustmentAn example using the window from the graph above:
Window deviations: +120, −180, +60, +640, −90, −140, +40
Balance: +450 kcal
Adjustment: −450 ÷ 7 = −64 kcal
Original goal: 2,000 kcal
Room for today: 1,936 kcalNotice that the +640 kcal Saturday turned into an adjustment of 64 kcal a day. Divide, don’t absorb. No single day tries to fix a whole week on its own.
Why there’s a 300 kcal cap
The ±300 kcal cap exists for two reasons, and both are about safety.
Downwards, it guarantees that no day turns into a starvation day. Even after a run of days well above goal, today’s target never drops by more than 300 kcal. The window handles the rest: the heavy days will leave the count on their own.
Upwards, it guarantees that a run of days below goal doesn’t become a licence for one day far above. Eating 300 kcal more than your goal is a more generous dinner, not a free-for-all meal.
Why today is left out
Today isn’t part of the average. It’s the result of it. That has a practical consequence you’ll notice right away: your room for today doesn’t move as you log the day’s meals. You wake up with a number and it stays put until tomorrow. What changes through the day is your progress towards it, not the target.
How this talks to adaptive TDEE
D-Fit has two mechanisms that look at your calorie average, and it’s worth understanding how they split the work.
Adaptive TDEE looks at a 28-day window. It cross-references your average logged calories with your weight trend and works out how much you actually burn. From that, it suggests a goal adjustment, once a week. It decides where the line sits.
The 7-day Balance looks at a 7-day window. It measures how far you are from the line and suggests your room for today. It helps you stay on the line.
The two never fight, for a simple reason: the balance measures deviations against each day’s own goal. When adaptive TDEE moves the goal, old days keep being compared against the old goal, and new days against the new one. No adjustment creates a fake debt.
Adaptive TDEE (28 days) → sets the goal
7-day Balance (7 days) → keeps you on the goal
Today → what you do with the roomHow to use it day to day
Watch the line, not the bar. A pink bar (a day above goal) isn’t a problem. A line that climbs for several days in a row is the signal that it’s worth adjusting.
Let the weekend be the weekend. If Saturday tends to be your highest day, eating slightly below goal from Thursday to Sunday morning already sorts it out. Orsama’s research shows that’s how the people who actually lose weight do it.
Don’t chase your room for today to the letter. It’s a guide, not a rule. If today says 1,936 and you finished on 1,980, the average will absorb that without drama. The 300 kcal cap already guarantees the guide never gets aggressive.
Log your off days. The average only works with data. A day with no log simply doesn’t enter the count, and the window gets smaller. Logging a 2,800 kcal day isn’t a confession, it’s the information that makes Tuesday’s number correct.
Use Monday as a landmark, if that helps you. The fresh start effect is real. Review the goal, look at the 7-day graph, adjust the plan. Just know the account didn’t clear. It never clears, and that’s why it works.
FAQ
Why 7 days and not 30?
With 30 days, what you eat today is 1/30 of the average. That’s too little for you to feel the effect of your choices and correct course. With 7 days, each day weighs enough to matter, and the window is still long enough to absorb a weekend. It’s the same reasoning that makes adaptive TDEE use 28 days: there the goal is to measure a stable burn, here it’s to give you fast feedback.
What happens if I go several days without logging?
Those days stay out of the average. If the window has fewer than 3 logged days, the app shows the average but doesn’t adjust today’s goal, because the information is thin. It will ask you to log a few more days before suggesting anything.
Does the average include exercise?
Each day’s goal is the same number you see in your diary. The balance compares what you ate against that goal and doesn’t add or subtract training calories on its own.
Does one day way above “ruin” the week?
There is no “the week” any more. A +640 kcal day becomes an adjustment of 64 kcal a day over the following seven days, and then leaves the count. If you’d rather not compensate at all, the average will absorb it on its own within a week, and your weight, as Hall’s model shows, will barely notice.
What if I still want to see the calendar week?
The 7-day graph shows the day of the week on each bar, so you can still see your weekly pattern. The only thing that changes is the math: it looks at the last seven days, not at “since Monday”.
Conclusion: a line, not a scoreboard
A diet isn’t a sequence of weeks that start and end. It’s an average that moves slowly, and what you do today only matters through what it does to that average. D-Fit’s 7-day Balance puts that on screen as a single line, and turns your job into a simple task: keep the line near zero, one day at a time.
It’s the same principle as flexible dieting and adaptive TDEE: measure what’s real, respond to what’s real, and leave the calendar to your schedule.
References
- Hall KD et al. Quantification of the effect of energy imbalance on bodyweight. The Lancet, 2011. doi:10.1016/S0140-6736(11)60812-X
- Harvie MN et al. The effects of intermittent or continuous energy restriction on weight loss and metabolic disease risk markers. International Journal of Obesity, 2011. doi:10.1038/ijo.2010.171
- Headland M et al. Weight-loss outcomes: a systematic review and meta-analysis of intermittent energy restriction trials lasting a minimum of 6 months. Nutrients, 2016. doi:10.3390/nu8060354
- Byrne NM et al. Intermittent energy restriction improves weight loss efficiency in obese men: the MATADOR study. International Journal of Obesity, 2018. doi:10.1038/ijo.2017.206
- Racette SB et al. Influence of weekend lifestyle patterns on body weight. Obesity, 2008. doi:10.1038/oby.2008.320
- Orsama AL et al. Weight rhythms: weight increases during weekends and decreases during weekdays. Obesity Facts, 2014. doi:10.1159/000356147
- Polivy J, Herman CP. Dieting and binging: a causal analysis. American Psychologist, 1985. doi:10.1037/0003-066X.40.2.193
- Polivy J, Herman CP. Overeating in restrained and unrestrained eaters. Frontiers in Nutrition, 2020. doi:10.3389/fnut.2020.00030
- Dai H, Milkman KL, Riis J. The fresh start effect: temporal landmarks motivate aspirational behavior. Management Science, 2014. doi:10.1287/mnsc.2014.1901
