Biohack #2: Energy levels, glucose, and afternoon slump

9 min readApr 9, 2021
Are you familiar with the afternoon slump?

My energy levels have improved a lot (both in terms of height and constancy) as I worked on fixing my sleep, exercise, diet, and mind. I wake up feeling relatively refreshed and I am able to remain productive all the way until lunchtime. The catch is — I still get an afternoon slump, which can be pretty nasty.

From about 2pm (immediately after lunch) to about 4, I struggle to remain focused and productive. I get tired and my brain goes foggy. I can push myself to keep doing menial tasks but this is definitely not the time for creative and high quality work. It is also not pleasant.

Second, about one hour before lunch and before dinner, I get the opposite of this — I start to feel anxious and shaky as my body starts reminding me that it’s nearing time to refuel. It feels different to the afternoon slump — instead of feeling like I’m being weighed down by a heavy burden — I feel like there’s a hole inside me that needs filling, and I struggle to refocus my mind on anything else than the completion of this task. I struggle to articulate clearly and my mind gets filled with a sense of urgency.

Now, according to various sources, energy levels fluctuate naturally due to circadian rhythms, and I know these changes can be exacerbated by things like elevated stress levels (which I’m addressing separately). All in all, there is nothing ‘abnormal’ about this state o affairs, however, that doesn’t mean things can’t and shouldn’t be better, with me feeling productive and on top of things, all day.

Hacking my energy levels constant

Let’s design a series of experiments, to establish what factors have an impact. According to the internet, some of the most common factors are —

  1. Sleep — currently my Oura tracker says my Sleep Balance (average amount of sleep over the last 7+ days) is slightly low. I will do my best to catch up on some sleep this weekend (even though my average has been about 7 hours / night, over the last 2 weeks) to try reset my baseline, and I imagine it will help a bit. That said, the slump is a long-term thing for me and I also want to get to a stage where I don’t face this slump even when I’m low on sleep, hence I will continue compiling this list.
  2. Snacking — an obvious solution. If I keep (healthy) snacking between meals, this should keep my sugar levels more constant and make my food craving less extreme, before meal time. I’m currently doing this and it has some effect, though not enough. Especially not in helping me reduce the afternoon slump.
  3. Hydration — a common recommendation and definitely a good idea. I am already drinking water at least every 2 hours but the effect is not noticeable.
  4. Carbohydrates — reducing the amount of carbs consumed (especially for lunch) seems to be an oft-recommended solution. It makes sense. Digesting carbs makes you tired and (similar to sugars), makes your blood sugar spike and then drop down.
  5. Gluten — apparently non-celiac gluten sensitivity is a relatively common thing, with symptoms including fatigue and brain fog. A few popular diets involve gluten elimination and there are plenty of people who swear by the effect this is having on their physical and mental performance.
  6. Supplements — there seems to always be an answer in supplements, especially when it comes to managing one’s energy levels. At current, I’m entirely supplement-free and although it is something I am interested in starting to explore, I believe the sensible approach is to do lifestyle changes > diet optimization > supplements, in this order. If I can optimize for a variable by addressing the root cause without it, there is no need in creating supplement dependency.


Running the experiment

  • I will get my sleep balance (1) in the best order possible before next week and keep snacking (2) and hydrating (3) about every 2 hours, as a baseline behaviour.
  • I am very interested in experimenting with gluten elimination (5) but it would involve some significant changes to my diet and take about 2 weeks for changes to be noticeable. Hence, I will start with running a series of experiments around carbohydrates (4) and glucose, throughout the next 2 weeks.

The problem is that I’m currently in a bulking phase and need to maintain a caloric excess — making it hard to cut on carbs and still hit my target. As per my previous article — I am currently powering my lunch by the nutritionally-complete Huel meal replacement. I will leverage this, by A/B testing three different scenarios —

Huel meal replacement. I normally take 800 Cal of this for lunch.

A. No change: Take 800 Cal of Huel Powder v 3.0 (the standard product) for lunch, at 1pm. This gives me 72g of carbs, 58g of protein, and 26g of fat.

B. Less carbs, more protein: Take 800 Cal of Huel Black Edition (33% more protein, 50% less carbs) for lunch, at 1pm. This gives me 36g of carbs, 80g of protein, and 34g of fat.

C. Split lunch into two: Stick to Huel Powder v 3.0 but split my lunch up to 400 Cal at 12pm and 400 Cal at 2pm.

Depending on the results, I may experiment more with different timings of (C) or try a combination of (B) and (C).

Blood sugar levels are instrumental in influencing your energy levels and moods. Minimizing and spacing the intake of carbs should result in more stability, throughout the day. Image courtesy of thedrjoe.

Key metrics

One of the biggest issues will be to find an objective way of quantifying the results. I can think of two things —

  1. Track my subjective sense of: Energy level (1–5), Mood (word descriptor), Restlessness, Stress, and Anxiety (1–5 each). I have added a few metrics which might or might not be as relevant, in order to get a more granular image of how I feel. I will be using the Bearable app to track these and I will make an entry each day at 12pm (pre-lunch slump onset), 3pm (peak afternoon slump), and 6pm (pre-dinner slump onset). This spaces out nicely every three hours.
  2. Track my sugar levels: In order to do this, I will need to get a glucose monitor. I have tried measuring my glucose levels before, with the result clearly showing I’m at healthy levels, but I have not paid attention to fluctuations within the day, in such a structured manner. I will take my readings at the same times I do my mood tracking — 12pm, 3pm, 6pm.



I’ve ended up ordering a FreeStyle Libre continuous glucose monitor, which I will attach to my body, to keep monitoring my glucose levels throughout the day. It will be much less hassle than to keep pricking my finger and I will get a much more granular view of how my body reacts to different foods, activities, and environments, throughout the whole day.

FreeStyle Libre continuous glucose monitor. The device + two sensors costs 205 USD. Each sensor should last up to two weeks.

Key findings after 4 weeks:

  • Generally speaking, the more carbs-heavy meals I eat, the more I spike
  • If I eat a fat-based breakfast instead of carbs, my glucose levels will be much more constant throughout the day (2 & 3), regardless of what I eat for lunch. My average reading for the whole day will be also lower (by up to 0.6 mmol/l).
  • If I take a short walk or exercise after a meal, my glucose will spike less or come down faster
  • If I’m stressed an my sleep quality is poor, I will observe a series of fluctuations throughout the night (1 vs 2)
  • Eating fat and protein heavy meals for the whole day and having carbs for dinner has resulted in the sharpest spike yet (3)
1–2–3. Some quick annotated readings from different days, showing very different patterns.

Fluctuations within a day

As you can see on the aggregate hourly chart below, my median glucose is fairly level throughout the day (between 4.3 and 5.3), with the highest deviation after breakfast (9–11 am) and dinner (7–9 pm). This makes sense, as these are the two heaviest meals of my day. The lowest median is around 5 am, two hours before I get up, with very little deviation.

Improvement over the 4 weeks of the experiment

On the chart below you can see how my daily median glucose level progressively became lower throughout the experiment. This is due to me gradually switching breakfast away from oats (high in carbs!) to greek yoghurt with nuts and seeds (almost entirely fats & protein), and lunch from the standard Huel 2.0 to Huel Black (lower carbs, high protein). I’ve stuck to my usual dinners fuelled by carbs, but the impact on my daily median glucose level has still been significant (day =5.2, day 28 = 4.4). The difference between these two numbers is the same as the deviation between the 25th and 75th percentile, on most days!

Impact on my mood and energy levels

Finally, what about all the subjective readings I have been tracking? I did not manage to be consistent enough in tracking my restlessness, stress, and anxiety (seems like it’s harder to be disciplined enough to do this than I though!), but I did manage to get almost all the data points on my energy levels.

The result is… beautiful when one looks at the numbers. There is a significant negative correlation (-0.47 on a 3-hour segment level and -0.62 on a daily level) between my average glucose level (as measured) and my energy level (as self-reported). The p value for this is 0.0002 so, in other words, this correlation is highly unlikely to be random.

In layman’s terms — the higher my glucose, the more tired I feel. Not so shocking, ha!

What’s next?

This experiment gave me a good understanding of how what I eat affects my glucose and how this, in turn, affects my energy levels, throughout the day. I have learnt how to switch to less carb-heavy meals and achieve some positive results.

I was also amazed how much information about my day gets encoded in this data, when I correlate it with other observations — about my exercise, stress levels, quality of sleep, which days I was out on a date and ate junkier dinners than what I made at home, etc. It’s a fascinating world of its own (and quite creepy, when you think about the level of surveillance possible!) No wonder Apple are working hard to integrate non-invasive glucose sensors into their watch, in order to provide better real-time health and fitness data.

Going forward, I will largely stick to the diet I have established now, though I want to experiment with more variety of ketogenic meals, other than the yoghurts and nuts I have been largely dependant on. I will also try to reduce the amount of carbs in my dinners too, down to the bare minimum and see what happens. I will continue my measurements for at least another month to monitor this.

I’ve realised I also have a wealth of excercise, hearth rate, stress, and sleep data from my Garmin, and calorie and food composition data from MyFitnessPal. I might try to establish some more advanced correlations between all these and my glucose, later on.


Read the summary of all my #biohacking efforts so far in this article.

To see more and track my progress, follow me on Instagram and Twitter.