In short
- A candle hides sequence. Two identical candles can be built from opposite stories.
- Organic moves broaden: more participants, spread across venues, size distribution stays normal.
- Manufactured moves narrow: repetitive sizes, one venue, thin book, no follow-through elsewhere.
- Waiting for a five-minute close throws away the only window where the two are distinguishable.
- The realistic payoff is avoiding a bad entry, not catching the top.
Let us be precise about the subject. This is about recognising manufactured volume so you are not the person buying its final minute. Organised pump-and-dump schemes are market manipulation, they are illegal in most jurisdictions, and the arithmetic of joining one late is unkind. The useful skill is telling them apart from real moves quickly.
What a candle throws away
A candle preserves four numbers and discards sequence. Consider two five-minute candles with identical open, high, low, close and volume:
- In the first, buying arrived steadily across the whole five minutes from many different-sized orders.
- In the second, nothing happened for four minutes, then a burst of near-identical orders in the last forty seconds took price straight up.
Same candle. Completely different events. Anyone reading only the chart cannot distinguish them; anyone reading the trade feed cannot confuse them.
The footprint of an organic move
Real demand tends to broaden. The tells, in rough order of reliability:
- Participant count rises with volume. More distinct trades, not just bigger ones.
- Size distribution stays ragged. Human and algorithmic order sizes are messy — 0.37, 1.2, 0.05. Real flow keeps looking messy as it accelerates.
- It appears on other venues. Genuine repricing propagates: arbitrage drags every book that lists the asset within seconds.
- Related pairs react. The same asset against BTC moves, not only against USDT.
- Pullbacks get bought. The first dip meets resting demand rather than a vacuum.
The footprint of a manufactured one
Manufactured moves tend to narrow. They have to: the whole exercise is to move price using as little capital as possible.
- Repetitive sizes. Rows of near-identical amounts, spaced at suspiciously regular intervals — the signature of a script.
- One venue leads and stays alone. Price moves on the thinnest listing and the deeper books do not follow, or follow only reluctantly via arbitrage.
- A book that was already thin. Check what the pair was doing an hour earlier. Manufactured moves start in illiquid markets because that is where they are affordable.
- Vertical price with unremarkable notional. A large percentage move that cost surprisingly little to produce.
- No absorption on the way up. Price climbs through levels that would normally have sellers, because those sellers are the ones running it.
Combine two or three of these and the picture is usually clear inside a minute or two. See buy/sell delta for the absorption part specifically.
The five-minute candle is the wrong clock
The window in which these two patterns are distinguishable is roughly the first sixty to ninety seconds. Wait for a five-minute close and you have thrown away the entire diagnostic period — and, on a manufactured move, you are often buying the exit liquidity the operators arranged the candle to attract.
This is the practical case for reacting to trade-level events rather than candle closes. Not because faster is inherently better, but because the information that separates the two cases only exists at that resolution.
What ninety seconds is actually worth
Be realistic about the payoff. Ninety seconds does not make you early enough to catch a top, and nobody should be building a strategy around it. What it buys is mostly defensive:
- Not entering. By far the most valuable outcome — recognising a move you should sit out.
- Managing an existing position. If you already hold the coin, knowing whether the spike is broad or narrow changes whether you tighten or hold.
- Cancelling stale orders. Resting limit orders near a manufactured spike get filled at exactly the wrong time.
What none of this tells you
The honest limits, again. This does not tell you where price goes next: organic moves fail all the time, and manufactured ones occasionally start something real. It does not identify anyone — the feed is anonymous. And it does not work on every pair; on the deepest markets, no single participant can produce this footprint at all, which is itself a useful thing to know.
What the flow gives you is a better-informed few minutes than the chart gives you. That is worth having, and it is all that is being claimed.
Common questions
Can you detect a crypto pump early?
You can often distinguish a manufactured move from an organic one within the first minute or two by looking at the trade feed: repetitive order sizes, a single thin venue leading, and no reaction on related pairs point to manufactured volume. That is recognition, not prediction — it does not tell you where the price goes next.
What is a pump and dump in crypto?
A coordinated scheme where participants buy a thinly traded asset to drive the price up, promote it to attract outside buyers, and sell into that demand. It is market manipulation and is illegal in most jurisdictions.
Does volume always move before price?
Volume and price move together — volume is what moves price. The practical point is that the trade feed reports it at the moment it happens, whereas a candle reports it only when the period closes, which can be several minutes later.
Why does a five-minute candle hide this?
A candle keeps open, high, low, close and total volume, and discards the order in which the trades arrived. Two candles that look identical can be built from a steady broad move and from a forty-second burst of identical orders.
Pharos turns this into a push
Set one rule — say, buys above 5 BTC on your favourites — and the app tells you how often it will fire before you save it. Then it pushes you the moment it does.
See how Pharos worksWritten by the team building Pharos Trade. We make an alerting tool, not a trading service — nothing here is investment advice, and none of it predicts a price. Figures in examples are illustrative. Last reviewed 2026-08-23.