Can You Gain Weight In A Calorie Deficit?

This is an article by Eric Helms and Lawrence Judd (now Chief Marketing Officer at Macros Inc) originally published on

What you’re in for:

  • ~4600 words
  • 15-30 minutes reading time

It was another dreaded Monday morning. Sarah hit the snooze button for the third time, before jolting awake with the sudden realisation that no – she wasn’t sitting on a beach in Thailand. She had coffee to drink (3 cups, because who the hell functions on a Monday morning with any less?), breakfast to wolf down and then a mountain of emails at the office to which she needed to write snarky replies to, delete said snarky replies and re-write more professional replies.

Sarah clawed her way out of the blanket cocoon she’d made, and staggered to the shower. Catching sight of herself in the mirror, her heart sank. In the dream, she’d looked svelte, sleek and athletic, with toned legs and stomach to die for. She pinched the fat around her hips and grimaced – something had to change.

On her lunch break, she strode to the gym she’d been a member of for a while, but had actually stepped foot in far fewer times than she’d care to admit.

It was time to hire a personal trainer.

12 weeks later, and Sarah was over the moon. She’d been educated about the importance of a calorie deficit, protein intake and regular weight training. She could now goblet squat half her own bodyweight, was well on the way to nailing her first set of unassisted pullups and loved the feeling of high rep hip thrusts.

She also no longer grimaced when she looked in the mirror – there was still some way to go, but gone were the love handles and wobbly inner thighs. She’d actually dropped a dress size, her friends had started to compliment her on her hard work, and her boyfriend was also very happy with her new-found self-esteem and body.

However, she was curious as to how her weight had changed over the 12 weeks. Her personal trainer had weighed her at the start of the process, but then instructed her to not weigh herself at all. Today, however, that would change – today was officially 12 weeks since she’d started, and her trainer wanted to assess progress.

As she stepped on the scales, she saw her trainer look confused. She looked down at the scales, and understood why – her weight had gone up by half a kilo. How could that have happened? She’d been tracking her calorie intake, she looked leaner, and yes – she’d built some muscle… but surely not that much?

I’m sure you’ll have encountered something like this before. Either you (or a client, if you’re a personal trainer) thinking you’re in a calorie deficit, but being able to build muscle and also gaining net weight over time, despite getting leaner.

What’s going on?

Misreporting calorie intake? Some strange “newbie gains” phenomenon?

Both of these are plausible, but there is another, far simpler explanation.

Energy Balance

This explanation revolves around energy balance.

The theory of energy balance (as relating to human bodyweight) states that changes in bodyweight are dependent mainly upon your calorie intake and calorie expenditure. This is based on one of the fundamental laws of the universe – the first law of thermodynamics.

The first law of thermodynamics revolves around the concept of energy conservation: energy cannot be created or destroyed – only converted from one form to another.

Note: This article isn’t going to be an in-depth exploration of whether or not energy balance applies to humans; that is not up for dispute. There are countless metabolic ward studies supporting the theory. Rather, it is going to be exploring some of the nuances of how energy balance can apply to humans – the “grey areas” that tend to throw most of us when our weight doesn’t move as predicted over time, especially when resistance training becomes a significant part of the equation.

I’m also going to use the phrases “energy surplus” and “energy deficit” rather than the more commonly used calorie surplus/deficit. For all intents and purposes, they’re one and the same (given that energy can be measured in calories) but it simply saves me from constantly swapping between the two words and potentially confusing some people. Okay? Good.

The first question to ask is this:

Why does bodyweight change when we are in an energy surplus or deficit?

Potential Energy, And The Simple Model Of Energy Balance

To begin answering this, we need to understand what is actually changing within the body when we’re talking about an energy surplus or deficit. We don’t have “calorie stores”, do we?

Well, as it turns out… yes. We do.

The diagram below shows the basics of a thermodynamic system. Examples of thermodynamic systems that are slightly more easily visualised than a dotted line around a grey blob are:

  • A Thermos flask or hot water bottle
  • A kettle
  • The human body

Thermodynamic system

Essentially – anything that has a clear boundary, which can also interact with its surroundings in a way that heat and matter can enter and leave the system, can be referred to as thermodynamic system.

In humans:

  • Matter enters and leaves via food and excrement (faeces, urine, sweat)
  • Heat is constantly exchanged with our surroundings due to the fact we generate heat (called thermogenesis)
  • We have a clear boundary between the “system” and the surroundings – our skin!

Heat and matter entering or leaving the system from the surroundings changes the “potential energy” of the system. That is – the energy of the system relative to its surroundings (there are some other factors involved when defining potential energy, but they’re not especially relevant here).

This is what is changing in our bodies when we talk about energy surplus or energy deficits – the potential energy of our bodies.

If you’re struggling with the concept of the energy being “potential”, a rather morbid example is the one of you being stuck in a locked room with a limitless supply of water, but no food. It is reasonable to assume that the heavier you are, the more potential energy you have, and thus the longer you’re likely to survive. You have the potential to live longer.

If you’re still thinking in calories, the potential energy of the body can be equated to how many calories you have access to in storage in your body.

In terms of how the potential energy of our human body changes:

Net energy deficits must lead to a net reduction in the potential energy of the body, and net energy surpluses must lead to a net increase in the potential energy of the body.

I’ve displayed this visually in the two images below. The size of the energy in/energy out arrows (not to scale) is my way of indicating an energy surplus or deficit.

Energy in vs energy out

How do we measure changes in potential energy of the human body? The most common estimation is changes in body weight. But Sarah’s story at the beginning of the article, and the experiences of many trainees worldwide, hints at weight not potentially telling the full story.


  • Energy balance is one of the fundamental laws of the universe, and cannot be violated
  • A human body is a thermodynamic system, and thus must abide by the laws of thermodynamics
  • Energy surpluses or deficits change the amount of potential energy stored by the human body
  • Body weight changes are maybe not always a good enough proxy measurement for changes in potential energy

Science 101 – How To Tackle This

We have a model that in most instances, works pretty well – there’s a whole heap of metabolic ward studies showing that energy surpluses and deficits reliably lead to weight loss and weight gain

However, we have some instances in which the model falls down. In this instance, it’s when resistance training forms a substantial part of the “energy out” side of the equation.

Therefore, we need a better model. To arrive at this better model, we need to think a bit like a scientist. This involves three steps:

  1. Identify some of the underlying assumptions of the current model that maybe aren’t valid
  2. Address and improve upon those assumptions
  3. Modify the model to include the improved assumptions

Assumption – The Homogeneity of The Human “System”

The immediately obvious assumption of this model is that the human body is homogeneous in its potential energy – that is, that every single bit of the human body contributes equally to the overall potential energy of the system.

This is clearly untrue – most of us are aware of the different calorific values of each of the macronutrients, so it isn’t too far a leap of faith to realise that our body fat stores may contain more potential energy per unit mass than our muscle stores. Different tissues (muscle, fat and organ tissues) have different potential energies – referred to as metabolizable energies.

Livesy and Elia estimated the metabolizable energy densities of glycogen, body protein and body fat back in 1988 and found them to be 17.6, 19.7, and 39.5 MJ/kg respectively. However, these values don’t necessarily apply when the body undergoes significant changes in stores of each of these due to energy surpluses or deficits. This is due to changes in the water content of the tissues, which will alter the energy density by effectively “diluting” or “concentrating” the energy density of each of the tissues. This is especially true for protein and glycogen, which are stored with much more water than fat.

The metabolizable energy densities for changes in body fat and body protein were defined in 2008 by Hall, and were estimated to be 39.5 MJ/kg for fat and 7.6 MJ/kg for protein. Converting this into calories gives us approximately 9400 kcal/kg of fat, and 1815 kcal/kg of protein.

Important points to note with how these metabolizable energies were defined by Hall:

  • Glycogen changes were negated – Hall recognised that glycogen and protein have similar energy densities, are stored with similar amounts of water, and that glycogen will account for only a very small quantity of changes in lean mass compared to protein given the finite glycogen storage capacity of the body.
  • The energy density for fat changes is associated with pure fat being metabolised – not a change in adipose tissue (body fat). Body fat does contain some water and some protein, found in the form of cellular machinery, which decrease the energy density of adipose tissue by a small amount. However, this is still an appropriate number to use as:
    • Human fat cells are filled and emptied rather than the cell being fully metabolised.
    • Most body composition estimation techniques count changes in water as changes in lean body mass rather than changes in fat mass. Fat mass and adipose tissue mass aren’t quite the same thing, but it’s virtually impossible to measure changes in adipose tissue.
  • The value for protein changes takes into account changes in the relative hydration of lean body mass during weight loss.
  • The eagle-eyed amongst you (and those who’ve been good scientists and read Hall’s paper I referenced above) will notice that this metabolizable energy density for protein is associated with protein losses. Building muscle is a much more energetically costly process – however, any increased energy requirements will be accounted for in your total daily energy expenditure; the “calories out” side of the equation, if you will.This last point is a little complicated. Let’s say you want to build 10 grams of muscle in a given day (a number I plucked out of thin air) and the energetic “cost” of building that muscle is 300 calories (another number that I whipped out of nowhere). This simply means that you’ll need to eat an extra 300 calories in addition to the calories that you’re going to store as new muscle tissue. The energy you expend building the muscle is balanced out by the energy consumed, and so the energy density of the changes in body protein will remain the same as if you were losing body protein. The maths holds.

In summary – a kilogram of fat metabolised will release approximately 5.2 times as many calories as a kilogram of lean body mass. And conversely, it requires approximately 5.2 times as many calories to make up 1 kg of fat mass than it does 1 kg of lean body mass. The reason we don’t store more as lean body mass is largely one of energy efficiency – it is less energetically costly to fill up fat cells than it is to synthesise new muscle tissue.

Assumption 2 – Body “Recomposition” Is Generally Insignificant, And Doesn’t Occur In Trained Individuals

That’s the rule of thumb, right? To build muscle, you need to be in an energy surplus, and to lose fat you need to be in an energy deficit. If you’re bulking, you need to accept the fat gains that come with it. If you’re losing fat, you’re probably going to be losing some muscle tissue too. To attempt to do the two simultaneously is stupid, and futile, and only happens in beginner lifters… right?!

In reality, it’s not that black and white. There are a number of studies in which subjects have been shown to:

  • Build muscle and lose fat at the same time
  • Build significant quantities of muscle with little-to-no increases in body fat – every weightlifter’s dream

Were they in an energy surplus, or an energy deficit? Let’s find out.

Summary table

*The Maltais study had 26 study subjects in total

Most of the data in this table, and the associated energy surplus/deficit calculations, were compiled or calculated by Eric Helms. N indicates the number of subjects in each group and LBM stands for lean body mass.

Delta Q represents the total energy surplus or deficit over the course of the study duration, averaged for each group (indicated by a positive or negative number respectively) as calculated by the difference in metabolizable energy. It represents the average daily surplus or deficit – the difference in potential energy between the average start and end points of the subjects, spread over the duration of the study – but doesn’t necessarily represent what happened day-to-day with each subject in terms of their actual energy intake vs expenditure.

This comes with a number of pretty serious limitations:

  • The average Delta Q would almost certainly have been different at different time points over the duration of each study. The subjects would have fluctuated in and out of surpluses and deficits around the average daily value, as per the sketch graph below.
    TDEE average graph
  • It’s also averaged over all the subjects in each group – this means that there were almost certainly individuals in each group who would have been in a net deficit, despite the average Delta Q being an energy surplus (and vice versa). This is a pretty big limitation, and I’d love to have been able to do some deeper analysis on the individual data. However, given the way statistics tends to work, there will likely be some people who are represented by these data.

Think of it this way – you’re meant to be in a calorie deficit, but you REALLY want a slice of cake one day that’s going to put you in a surplus for that day. You decide to subtract calories from the remaining days that week so you remain in a deficit over the course of that week. It’s the same principle here, just measured over the duration of the study rather than a week.

In the table, we have data from 5 studies – Antonio et al in 2014 and 2015, Campbell et al in 2015, Maltais et al in 2015 and Verdijk et al in 2009 -totalling nearly 250 subjects, with a huge range of ages, training status and protein intake (and some with sarcopenia, which is a clinical condition resulting in muscle wasting and loss of strength with age).

We can pull the following observations out of this data:

  • All groups showed a significant increase in muscle mass – even both groups in the 2014 Antonio study, who were instructed to not alter their training throughout the duration – regardless of being in a net energy surplus or deficit by the end of the study. For those of you looking at the Verdijk study and thinking that 0.7 kg in 12 weeks isn’t much: these were people in their 70s, eating just over 1 g/kg of protein per day and doing 4 sets each of leg press and leg extension, 3 times per week. That’s good going!
  • 2 of the groups who were in a surplus actually lost body fat. These were very small energy surpluses (20-30 kcal over the course of the study), but happened in trained individuals and untrained individuals alike.
  • 6 of the groups that were in a deficit either gained small amounts of weight or were virtually weight stable.
  • The 3.4 g/kg group in the Antonio study only lost a total of 0.2 kg, despite a deficit of nearly 240 kcal.

It’s worth noting that the only subjects who were classed as trained individuals were the 4 groups in the Antonio studies, and many of the remaining subjects were overweight. This makes the Antonio groups even more surprising – they displayed changes in LBM just as large as a lot of the untrained individuals, over a shorter period of time, and the 3.4 g/kg group also showed the largest single drop in body fat over the study duration.

If the data are graphed, and trendlines/R­­2 values are plotted (to investigate whether there are correlations, and determine the strength of correlations should they exist), we can see some interesting data.


It should be noted that the correlations are all relatively weak, and would likely not reach statistical significance, however hopefully future research will help elucidate some of these observations:

  • Protein intake correlated relatively well with changes in muscle mass, but not with fat mass.
  • The Delta Q correlated best with fat mass (FM) changes (the strongest correlation out of all observed),
    then total weight change. It didn’t correlate at all well with gains in LBM.
  • Protein intake did not seem to influence Delta Q in any way, although digging deeper into the Antonio studies seemed to indicate that increasing protein intake meant that study subjects reported calorie intakes that were closer to what would be expected for their weight and activity level. Whether this due to the protein, or simply that they realised they were the intervention group (and so made more of an effort to track accurately) is impossible to say.

The surplus correlating with FM gains and total weight change but not LBM gains is a very interesting observation. This was also observed (albeit on a small scale) in elite athletes by Garthe et al, who added 4 strength-training sessions per week to 18 athletes’ regimens for 8-12 weeks. 9 of them were put in an energy surplus (~3500 kcal per day) and the other 9 had an ad-lib approach, averaging ~3000 kcal per day. The ad lib group gained just as much LBM as the group in an energy surplus, but only increased their FM by 3% vs 15% in the surplus group. This agrees with what Antonio et al observed in their 2015 study – that people (even trained individuals) will build muscle just by following an ad lib diet/their baseline dietary habits, and focusing on progress in the gym.

This fits nicely with what we see in practice – everyone from beginner trainees, who can make remarkable progress just from getting in the gym and lifting, to high-level natural bodybuilders, who seem to do well by following a “gaintaining” approach with very small calorie surpluses.

Updating The Model/Practical Take Home Points

We now have enough information to update the model slightly for when protein intake is sufficient and some intense resistance training is being performed regularly. Quantitative models for weight change taking into account activity have already been developed, and it’s a little outside of the scope of this article to try and do the same to account for weight training. Also, I might lose the only reader I have left by this point.

It seems more prudent to give some overarching summary points with the aim of managing expectations for people who want to eat lots of steak and lift heavy things:

  • When cutting, weight loss is still likely to be the most reliable indicator of progress. It’s also the most tangible – it’s easy to step on some scales, and they are far less invasive than attempting to measure your body fat regularly.
  • When bulking, weight gain can be much slower and decent progress can still be made. Remember, nutrition is permissive, and muscle growth cannot be force fed – the Garthe study shows this quite nicely, and this video by Eric explains the concept nicely too.
  • A weight stall when cutting doesn’t necessarily mean your progress has stalled, especially in novice or intermediate lifters and even more so in those who are training efficiently to maximise muscular hypertrophy.
  • Progressively increasing your training volume and eating sufficient quantities of protein is most likely to lead to favourable body composition outcomes.
  • This may add to the evidence explaining why bodybuilders who “eat up into a show” and still manage to lose weight overall whilst still increasing food intake.
  • The “gaintaining” approach is a perfectly valid approach. It may be especially useful for:
    • People who struggle with feeling “fluffy”, although the likelihood is that a small amount of fat will be gained.
    • People competing in weight-class regulated sports who want to maximise muscle mass at a given weight.
    • High-level intermediate natural lifters and beyond, who anecdotally seem to respond badly to aggressive bulking.
  • If you use the 3500 kcal/lb rule to predict the energy surplus or deficit associated with the weight change found in these subjects, it actually tends to give numbers that correlate quite well with the numbers calculated from the metabolizable energy densities (within 200 kcal, for the most part).From a practical standpoint, this means that for a lot of trainees who are weight stable, aren’t coming back to training from a layoff, aren’t undertaking a crazy new training regimen, or are enhanced lifters about to start a drug regimen (a.k.a all the scenarios in which rapid muscle growth is expected), the 3500 kcal/lb rule predicts weight loss quite well.It also predicts muscle gain quite well in those who are in similar scenarios to those listed above. Based on the metabolizable energy density, a pound of muscle contains approximately 800 kcal. Literature shows that it takes approximately 4-6 times as much energy to build muscle, which puts the total caloric “cost” of a pound of muscle at roughly 3500 kcal. In a practical realm, this means that setting up calorie intake for slow weight gain (gains of <1.5% of bodyweight per month) based on the 3500 kcal/lb rule should predict weight gain relatively well.


The main thing we want you, dear reader, to take home from this article is this – recomposition is normal. It happens – less so in trained individuals, but much more in untrained, new lifters. It’s not necessarily something to aim for, but be aware of it and how it can affect your scale weight if you use that as a metric to gauge progress for either you or your clients.


I (Lawrence) would like to tip my hat to:

  • Eric Helms for his insight, invaluable help with the initial data analysis and for doing me the honour of being a co-author on this article.
  • James Krieger, Ian McCarthy and Menno Henselmans for their hugely interesting and informative discussions on Facebook, which have proven very useful in writing this.
  • My fellow coaches at Shredded By Science for their feedback.
  • You, dear reader – for sticking with me this far.


  1. Wayne W Campbell, *. J.-H. (2015, Aug). Higher Total Protein Intake and Change in Total Protein Intake Affect Body Composition but Not Metabolic Syndrome Indexes in Middle-Aged Overweight and Obese Adults Who Perform Resistance and Aerobic Exercise for 36 Weeks. The Journal of Nutrition .
  2. Dr Kevin D Hall, P. G. (2011). Quantification of the effect of energy imbalance on bodyweight. The Lancet , 378 (9793), 826-837.
  3. Garthe I, R. T.-B. (2013). Effect of nutritional intervention on body composition and performance in elite athletes. . Eur J Sport Sci. , 13 (3), 295-303.
  4. Geoffrey Livesey, P. a. (1988). Estimation of energy expenditure, net carbohydrate utilization, and net fat oxidation and synthesis by indirect calorimetry: evaluation of errors with special reference to the detailed composition of . Am J Clin Nutr (47), 608-28.
  5. Hall, K. (2008). What is the Required Energy Deficit per unit Weight Loss? Int J Obes (Lond) , 32 (3), 573-576.
  6. Jose Antonio*, A. E. (2015). A high protein diet (3.4 g/kg/d) combined with a heavy resistance training program improves body composition in healthy trained men and women – a follow-up investigation. JISSN , 12 (39).
  7. Jose Antonio*, C. A. (2014). The effects of consuming a high protein diet (4.4 g/kg/d) on body composition in resistance-trained individuals. JISSN , 11 (19).
  8. Lex B Verdijk, R. A. (2009). Protein supplementation before and after exercise does not further augment skeletal muscle hypertrophy after resistance training in elderly men. American Society for Nutrition , 89 (2), 608-616.
  9. Leone PA, G. D. (2000). Relative overhydration of fat-free mass in postobese versus never-obese subjects. Ann N Y Acad Sci. (904), 514-519.
  10. Maltais ML, P. K.-L. (2015, Aug). Effect of Resistance Training and Various Sources of Protein Supplementation on Body Fat Mass and Metabolic Profile in Sarcopenic Overweight Elderly Men: A Pilot Study. Int J Sport Nutr Exerc Metab , epub ahead of print.
  11. Marion J. Franz, M. R. (2007). Weight-Loss Outcomes: A Systematic Review and Meta-Analysis of Weight-Loss Clinical Trials with a Minimum 1-Year Follow-Up. Journal of the Academy of Nutrition and Dietetics , 107 (10), 1755-1767.
  12. Williams, M.H., (2005) Nutrition for health, fitness, & sport, McGraw-Hill Science Engineering.


Submit a Comment

Your email address will not be published. Required fields are marked *