Blog
Blog
Visual temporarily unavailable
This section's visual is under review and a replacement is being prepared.
The orderbook looks like information. Most of what you see is theater.
Executed flow shows whether buyers or sellers are actually taking liquidity. That matters before price has fully reacted. What matters on the page is simpler: the visible book can mislead, while executed aggression leaves a harder-to-fake trace.
Reading the orderbook is not about finding hidden signals. It is about knowing which layer of it to believe.
The orderbook shows resting limit orders. Bids on one side, asks on the other. What most traders underestimate is how little of that resting liquidity is real in any durable sense.
A large order on the bid does not mean someone will buy. It means someone placed that order. It will disappear the moment it becomes inconvenient. Market makers cancel and reprice faster than you can blink. Algorithms place and pull orders as a probe, not a commitment.
The market rewrites that book faster than you can read it.
What the orderbook genuinely reveals is shape: where liquidity clusters, where it thins, and how symmetric or asymmetric the two sides are. Whether that shape is true pressure or deliberate misdirection is the harder question.
Resting limit orders can be faked. Executed trades cannot.
When a buyer crosses the spread and hits an ask, that trade is permanent. The aggressor paid the higher price to get filled immediately. Every market order, every marketable limit, every aggressive fill is a signal that someone wanted that position more than they wanted a better price.
Accumulate enough of those signals in one direction and you have a measurable imbalance. This is what traders who move serious size watch instead of the resting book. The resting book is decoration. The flow is the argument. The question is why anyone with serious size would bother painting the decoration.
The practical implication: if you see a wall of bids and the executed flow is still skewed to the sell side, buyers are advertising confidence they do not have. Price is more likely to test lower than to bounce. The wall is a prop.
Large participants place large visible limit orders they never intend to fill, to create the impression of demand at a level without actually committing to it. A large resting order is not evidence of intent to trade. It is evidence that someone placed an order.
One pattern worth knowing: a very large bid appears. Price drifts up slightly. The bid vanishes before price reaches it. Price then drops. The bid was not support. It was bait. A liquid pair generates several hundred to several thousand orderbook events per second during active sessions. Most traders are watching a one-minute candle.
Real buying pressure and fake buying pressure look identical on the resting book. They do not look identical in the flow.
Genuine accumulation shows up as persistent buy-side aggression: marketable orders consistently lifting the ask, small orders that absorb each dip in the offer, no large visible bids that disappear before execution. The book may look thin. The flow tells you buyers are serious.
Fake pressure shows the opposite signature: large resting bids that never get hit, sell-side flow continuing to press, the wall retreating as price approaches it. The chart looks supported. The flow says it is not.
Price catches up to flow. Not on a fixed schedule, but the gap between pressure and price is not random either.
Here is the part most traders miss. Orderbook pressure does not move price instantly. There is a lag, and that lag is not random.
Buy-side aggression builds for 30, sometimes 60 seconds before price responds. During that window, the chart looks flat. To anyone watching flow, the argument has already been made. The candle is simply waiting for the rest of the market to agree. Flow diverged from the resting book before price moved. The candle did not move. The argument was already over.
That lag is not an exploitable arbitrage. It is not a guaranteed trade. What it is: a window where the directional argument has been made before it shows up in the layer most traders watch. Confirmation is safer. Early is more interesting.
You watch buy-side aggression tick up. Then again. Then again. The candle does not move. Forty seconds of the same argument, the same side pressing, no response from price. You are reading conviction accumulate in real time, before the chart has any idea it is happening. That is the window. It is not a prediction. It is a measurement of pressure that has not yet resolved into price.
Even interpreted correctly, the orderbook has a hard limit: it only shows the current moment on the current venue you are watching.
Financial markets fragment across many venues simultaneously. When they disagree on price, one of them is about to move. A flow reading on a single exchange is a local measurement. This is not an edge case. It is the baseline condition of modern markets. The same fragmentation exists across equity dark pools, FX ECNs, and futures exchanges.
The other limit: context. A balanced orderbook with equal buy and sell flow is not a neutral signal. Flow tells you aggression. It does not tell you motive. Motive requires context: what happened in the preceding hour, what event is approaching, whether the asset has been trending or ranging for the last week.
In news-driven markets, all of this breaks. When a central bank announcement hits, executed flow means nothing because everyone is repositioning simultaneously. The signal is overwhelmed by the event. Knowing when to ignore it is as important as knowing when to use it.
Before any entry, one question: is the flow confirming what the chart suggests, or contradicting it?
When price is near a support level, the chart is showing you a line. Flow shows you what people are actually doing. If buy-side aggression is rising as price approaches that level, the support has backing. If sell-side flow is still pressing and the wall of bids is not getting hit, the level is aesthetic, not structural. The chart drew a line. The market did not agree to honor it.
The same logic applies on the short side. Resistance without sell-side flow is a ceiling no one is pushing against. A resting offer that retreats before price touches it was never real resistance. It was a signal someone was reading you.
None of this is a system. The market does not stop performing for you once you know it is performing.
The theater does not end. That is the point.
Every participant in the orderbook has a motive, a position, and a reason to mislead. Executed flow is the only layer that cannot be faked after the fact. It is the record of what participants actually did, not what they wanted you to think they would do.
Most traders watch the resting book their entire career. They read the script while the real action happens in the trades that have already executed. The chart shows you where price went. Flow shows you where conviction was, before the price moved to confirm it.
The orderbook looks like information. The part that is not theater is small, specific, and arrives before the candle does.
The bid wall was there when you checked. It is gone now. The executed flow never paused. That gap between what the book displayed and what the trades confirmed is the only signal that was real, and it was telling you the answer the whole time.
It shows resting bids and asks at different price levels. That is useful for seeing liquidity shape, but it does not prove those orders will stay in place or trade when price reaches them.
Because they can be canceled, repriced, or placed to create a false impression of support or resistance. That is why spoofing risk and fake walls matter when you interpret visible depth.
Executed flow. Once an aggressive order crosses the spread and trades, that event is real. This is the logic behind order flow imbalance and other execution-layer signals.
No. The better approach is to read book shape and executed flow together. The book gives context. The flow tells you whether that context is being confirmed or contradicted in real time.
It cannot tell you whether visible size is genuine, whether another venue disagrees, or whether current pressure will persist after the next wave of orders arrives. It is one layer, not the whole answer.
Cont, R., Kukanov, A., & Stoikov, S. (2014). The Price Impact of Order Book Events. Journal of Financial Econometrics, 12(1), 47-88. https://academic.oup.com/jfec/article-abstract/12/1/47/816163
Kyle, A. S. (1985). Continuous Auctions and Insider Trading. Econometrica, 53(6), 1315-1335. https://www.econometricsociety.org/publications/econometrica/1985/11/01/continuous-auctions-and-insider-trading
Order flow data across multiple exchanges, in one place.
No card required.