Why Do I Crash at 3pm?
Published · 8 min read
Because it is scheduled. The afternoon dip is a real, measurable feature of the human day that shows up even when you have not eaten lunch and do not know what time it is. Lunch can deepen it. Lunch does not cause it.
That distinction matters more than it sounds, and it is the whole reason the 3pm crash is worth a guide of its own. A slump caused by what you ate is a thing you react to. A slump that runs on a clock is a thing you can see coming.
It is not the sandwich
The clearest summary comes from Timothy Monk's review The post-lunch dip in performance in Clinics in Sports Medicine, 2005 (DOI 10.1016/j.csm.2004.12.002). His description is worth quoting closely: the dip is a real phenomenon that can occur even when the individual has had no lunch and is unaware of the time of day.
Two experiments make that concrete.
Lunch removed entirely. Monk, Buysse, Reynolds and Kupfer, Circadian determinants of the postlunch dip in performance, Chronobiology International, 1996 (DOI 10.3109/07420529609037076) tested people every two hours on a monotonous vigilance task with lunch replaced by hourly liquid food supplements. There was no lunch to blame. Five of the twelve participants still showed a clear dip.
Lunch moved. Javierre and colleagues in Revista Espanola de Fisiologia, 1996 (PubMed 9144846) had eight sprinters run 80 metre sprints eight times a day across five Saturdays, shifting the sleep and meal schedule between days. A dip appeared on every testing day. On the day the sleep cycle moved but mealtimes did not, performance started deteriorating at 15:00 while lunch was not served until 16:00. The dip tracked the body clock, not the plate.
Their conclusion was careful and worth keeping: the dip is not directly linked to the time of lunch, although lunch itself can potentiate its effects. They also noted that no other meal of the day does this.
Why some people feel it and others do not
The 1996 Chronobiology study is the interesting one here, because it compared the five people who dipped against the seven who did not.
| Group | What their body temperature rhythm looked like |
|---|---|
| Dippers | A higher amplitude, later peaking 12 hour component, leaving temperature flat from roughly 10:00 to 15:00 |
| Non-dippers | Temperature rising across the same window |
The difference could not be explained by how much they had slept, by whether they were morning or evening types, or by gender. The authors describe the dip as an endogenous phenomenon, individually determined, and linked to the strength of the 12 hour harmonic in the circadian system.
Read that again, because it is the practical part. Whether you crash in the afternoon is substantially a property of you, not of your discipline or your lunch order. Monk's later review adds that a high carbohydrate lunch does make it worse, and that extreme morning types may be more prone to it.
If you are trying to work out what is yours and what is circumstance, what is a personal health baseline covers the general version of that problem.
What actually helps, including what does not
Two interventions have been tested directly against the dip. The honest answer is that one has better evidence than the other, and that the weaker one is the one people reach for.
A short nap, taken before the dip rather than during it. Hayashi and Hori, The effects of a 20-min nap before post-lunch dip, Psychiatry and Clinical Neurosciences, 1998 (DOI 10.1111/j.1440-1819.1998.tb01031.x) put subjects to bed at 12:20 for twenty minutes and measured them every twenty minutes from 10:00 to 18:00. The nap improved subjective sleepiness and self rated task performance, and suppressed EEG alpha activity. Note the timing. It was taken at 12:20, ahead of the dip, not in the middle of it.
Bright light, where the result is genuinely mixed. Zhou and colleagues in Frontiers in Public Health, 2021 (DOI 10.3389/fpubh.2021.652849) took seventeen habitual nappers and removed the nap. Blue enriched bright light relieved some of the damage: sleepiness, negative mood and one working memory measure improved against normal indoor light. Sustained attention and inhibitory control did not.
And Askaripoor and colleagues in Industrial Health, 2019 (DOI 10.2486/indhealth.2018-0030) ran twenty volunteers through four lighting conditions and found something you should weigh before buying a lamp. Both red and blue enriched light improved the EEG correlates of alertness. Neither improved subjective sleepiness or task performance on any of sustained attention, working memory or inhibitory capacity. The brain waves moved. The work did not.
That is a null result and it belongs in this guide as plainly as the positive ones. A measurable change in a physiological correlate is not the same thing as getting more done.
One more, for anyone who schedules a drink at lunch. Horne and Gibbons, Ergonomics, 1991 (DOI 10.1080/00140139108967289) gave the same alcohol doses at 13:00 and at 18:30. Two and four units produced greater impairment in the afternoon than the identical dose in the evening. The same drink is not the same drink at the same time.
The thing that would actually change your day
Every study above measured the dip after it arrived. That is the right way to do science and the wrong way to run a Tuesday.
Consider what you do with each of these two sentences.
| What you are told | When you get it | What you can do |
|---|---|---|
| You crashed at three | Six in the evening | Write it down |
| Three is coming | Eleven in the morning | Move the meeting |
Same fact. One of them is a record and one of them is a plan. A dashboard that reports your afternoon is accurate and arrives too late to change the afternoon it is reporting on. Most tracking is built to file the first sentence.
The research above says the second sentence is possible, because the dip is endogenous, individually determined and tied to a clock rather than a sandwich. Something that recurs on a schedule is something that can be anticipated.
What to do with this
- Stop treating it as a failure of willpower. For a substantial share of people it is a property of their circadian system, and the evidence that it survives having no lunch at all is direct.
- Do not put your hardest work at your worst hour. If your dip is reliable, the cheapest intervention available to you is the calendar, not a supplement.
- If you nap, nap early. The evidence is for twenty minutes taken ahead of the dip.
- Be sceptical of the light claim. One study found real benefit on some measures, another found the brain waves moved and the performance did not.
- Watch the carbohydrate load, not the fact of eating. Lunch is a potentiator, not a cause.
- Find out whether you are a dipper at all. Five of twelve were, in the study that removed lunch. Population averages will not tell you which group you are in.
The limits, stated plainly
- The lunch removal study is twelve people, five of whom dipped. It is a careful design and a small sample.
- The sprinting study is eight athletes and measures one physical task. It is good evidence about timing and weak evidence about office work.
- The nap study measured subjective sleepiness and self rated performance, which are not the same as objective output.
- The two light studies disagree, and this guide has not resolved that for you. It has told you they disagree.
- Nothing here is medical advice. Persistent daytime sleepiness that is severe, new, or getting worse is worth taking to a clinician, and no tracker is a diagnosis. If your tiredness does not fit the pattern described here, why do I feel tired when my recovery score is good covers a different failure and may fit better.
The short version
The afternoon crash is on a timetable. It shows up when lunch is replaced with liquid supplements, it moves with the sleep cycle rather than the meal, it appeared at 15:00 in one study on a day lunch was not served until 16:00, and whether you get it at all appears to be a stable property of your own circadian rhythm rather than a verdict on your day.
Which means the useful version of that information is not a report at six in the evening. It is a warning at eleven in the morning.
That is the difference NUVARD is built around. It reads your own twenty four hour energy curve and tells you what is coming rather than what happened, so the hour that needs you most is not the hour you have already spent. Where something has no measurable effect on you, it says so, because that is a finding and not a gap.
Sources
- Monk TH. The post-lunch dip in performance. Clinics in Sports Medicine, 2005;24(2):e15-23. DOI 10.1016/j.csm.2004.12.002
- Monk TH, Buysse DJ, Reynolds CF, Kupfer DJ. Circadian determinants of the postlunch dip in performance. Chronobiology International, 1996;13(2):123-133. DOI 10.3109/07420529609037076
- Javierre C, Ventura JL, Segura R, Calvo M, Garrido E. Is the post-lunch dip in sprinting performance associated with the timing of food ingestion? Revista Espanola de Fisiologia, 1996;52(4):247-253. PubMed 9144846
- Hayashi M, Hori T. The effects of a 20-min nap before post-lunch dip. Psychiatry and Clinical Neurosciences, 1998;52(2):203-204. DOI 10.1111/j.1440-1819.1998.tb01031.x
- Zhou Y, Chen Q, Luo X, Li L, Ru T, Zhou G. Does bright light counteract the post-lunch dip in subjective states and cognitive performance among undergraduate students? Frontiers in Public Health, 2021;9:652849. DOI 10.3389/fpubh.2021.652849
- Askaripoor T, Motamedzade M, Golmohammadi R, Farhadian M, Babamiri M, Samavati M. Effects of light intervention on alertness and mental performance during the post-lunch dip: a multi-measure study. Industrial Health, 2019;57(4):511-524. DOI 10.2486/indhealth.2018-0030
- Horne JA, Gibbons H. Effects on vigilance performance and sleepiness of alcohol given in the early afternoon versus early evening. Ergonomics, 1991;34(1):67-77. DOI 10.1080/00140139108967289
Article records retrieved from PubMed.