The plateau, defined

Most "plateaus" are not plateaus — they are weather: water weight, glycogen swings, and one bad weekend wearing a statistician's costume. This page draws the exact line between a genuine stall (which deserves a response) and noise (which deserves patience), then hands you the measurement protocol that tells the difference in about two weeks of daily weigh-ins.

The Definition That Earns Action

Here is the working definition used across this series, and it is deliberately boring: a plateau is two full weeks with no downward trend in the weekly average weight. Not one flat week. Not five flat days. Two complete seven-day averages that fail to decline, read on the trend line, not on any single morning. That threshold exists for a reason: it is roughly the window where the ordinary water-weight swings of the next section systematically cancel out, leaving only signal. Anything shorter than that is below the resolution of your measurement — and responding to it makes things worse, not better.

The second test is a waist measurement the same morning. If the scale has sat flat for two weeks but the waist has moved down a centimetre, you are likely recomposing — losing fat while holding or gaining lean mass — and the stall is partly a costume. The Body Composition topic owns the measurement details; the rule here is that a flat scale plus a shrinking waist is good news wearing bad news clothing.

The logic behind the two-week line is worth stating plainly because it is a statistics problem, not a biology one. Fat loss on a sensible deficit proceeds at roughly half a kilogram a week at best — a few tenths of a percent of body weight — while ordinary daily water swings run a full kilogram or more in both directions. On any given morning the noise is several times larger than the signal you are hunting. Averaging seven days cuts the noise roughly by two-thirds, and comparing two consecutive weekly averages cuts it again. Two weeks is simply the shortest window where the trend can show itself above the static; anything less, and you are reading weather forecast for a specific afternoon.

What a Plateau Is Not

Half the job of defining a stall is listing what does not qualify. Each of these produces a flat or rising scale without a single extra gram of body fat behind it:

• 🚫 Five flat days. Within the range of ordinary noise. The scale is a slow instrument; a week is its minimum useful sample.
• 🚫 One heavy weekend. A restaurant weekend can add a kilo or two of water and food mass that takes days to clear — not a trend.
• 🚫 Three days without a new low. New lows arrive in bursts between plateaus; their absence for days is not information.
• 🚫 A scale that moved up 1–2 kg overnight. That is water and glycogen, period. Fat cannot appear that fast.
• 🚫 The week after a hard new training block. Repair-related fluid retention can mask real fat loss for weeks — the waist and the clothes are the tiebreaker.

Where the Water Comes From

To read a stall honestly you need to know the size of the noise you are trying to see through. These are the usual suspects, ordered by how much scale weight each can add — and every one of them is water, not fat:

Glycogen deserves a sentence of its own because it is the most common misread. Every gram of stored carbohydrate binds roughly three grams of water (Trexler et al., Journal of the International Society of Sports Nutrition, 2014). A single high-carb day after weeks of cutting reloads the stores, and the scale celebrates with a jump that is almost entirely fluid. It flushes over the following days — which is why judging any of this on a single morning is a fool's errand.

Sodium deserves the second sentence. Salt pulls water out of the bloodstream and holds it in the tissues; a restaurant-heavy weekend can leave you carrying a litre or two of fluid for days, and the flush arrives long after the memory of the meal. The practical symmetry is that the same mechanisms cut the other way: a few disciplined days of lower sodium and moderate carbs produce a satisfying "whoosh" of a kilogram or two — water leaving, not fat gained or lost. Neither direction is a verdict about your diet. That is exactly why this page insists on the fourteen-day average before any response: it is the one number these mechanisms cannot fake.

The Measurement Protocol

A stall verdict is only as good as the data behind it, so the protocol comes first. Run it every morning for two weeks before you are allowed to use the word "plateau":

• 🥱 Weigh right after waking, after the bathroom, before food or drink. Same scale, same spot, roughly the same time. Conditions are everything.
• 📏 Measure the waist the same morning. Same spot — for most people, at the navel — and log it with the weight.
• 🧮 Compute the seven-day average, not the day's number. Monday's figure is noise; the mean of the last seven is signal.
• 🔁 Compare week one's average to week two's. Flat average across fourteen days plus a flat waist: real stall. Flat average plus a shrinking waist: recomposition.
• 📝 Note the optional second channel. How the waistband sits and whether the ring is loose are free, daily, and remarkably honest — write the trend down even when it is awkward.

The Stall Decision Rule

Put the pieces together and the decision table is short. Judge on the fourteen-day average; everything else is detail:

Notice the pattern: only one row of four earns a response, and even that response is an audit, not a crash. The whole point of defining the plateau precisely is that the definition is what protects you from overreacting to the three rows that are not stalls at all.

Practical Rules for Not Fooling Yourself

Questions, Answered Briefly

• ⏳ How long do I really have to wait? Fourteen days of weekly averages is the honest answer. A week is too short to clear water noise; three weeks risks wasted time when the stall is real. Two is the balance.
• 🌊 I lost two kilos overnight — is that good? It is water and glycogen leaving, not fat at that speed. Celebrate quietly and keep the average as your scoreboard.
• 🩸 Does the monthly cycle really mess with the scale? Yes — predictable multi-day fluid swings are normal and expected. Schedule around them: judge the trend, never the morning-of.
• 📏 My waist is shrinking but the scale is flat. What now? That is recomposition — the best possible outcome of these two. Change nothing and recheck in two weeks; the flatline is your water layer arguing with the tape measure.
• ❓ What if I've been flat for six weeks? Then you are past "defined" and into the audit — Subtopic 3 in this series is the systematic checklist, and it starts by assuming intake drift, not a broken metabolism.

Related Topics