Smart Money Concepts II — TraderzLife Education Series Book 20

LEVEL 6 · ADVANCED · BOOK 20

The interlocking half. If the first book weighs the framework against the evidence, this one goes underneath the chart and describes the market the framework is trying to describe: who is actually trading, how a large position really enters, what order flow is and is not, and why almost none of it ever reaches your screen.

It does not tell you that the institutions are hiding from you. It shows you the machinery, and the machinery turns out to be stranger and far less personal than the story.

The one claim it makes: order flow works precisely because it is private. By the time anything about it is visible to you, it is already in the price.

• $9.6 trillion changes hands on an average day. Spot — the thing your chart shows — is 30.8% of it. The largest single instrument is the FX swap, at 41.5%, and it has no chart at all.
• Dealers in the major trading hubs match more than 80% of customer trades inside their own books, never sending them to a venue. No print, no volume, no candle. Their own description of what that achieves is worth the chapter: it reduces information leakage.
• There is no tape — and the book is precise about what that does and does not mean. Aggregates are published, but they are estimates: a three-yearly survey of more than 1,100 dealers across 52 jurisdictions, six regional committee surveys that cannot be added together because they count by different things, and individual venues reporting only their own flow. More than seventy-five venues exist. The central bankers' own phrase for the market is that it is crucial and inherently opaque.
• Order flow really does explain the price. In the study that started the field, flow and the interest differential together account for about 64% of daily variation in one pair — against under 1% for the interest differential on its own. But the forecast needs the day's realised flow, which you do not have until the day is over, and the authors say so themselves in a parenthesis the published version dropped.
• The relationship is a plateau with a peak, not a slide: strong from one minute right through the week, strongest around daily, fading only at a month and beyond.
• Large orders are sliced by execution algorithms whose published purpose is to avoid leaving distinct patterns that pattern-recognition can find. Four widely repeated beliefs about how they work — about the benchmark, about stealth, about adaptiveness, and about what slippage even means — are wrong, and the last one is wrong in a way that costs money.
• Impact does not round-trip. Across three distinct datasets and four independent analyses, one of them over eight million large orders, price relaxes to roughly two thirds of peak by the close and then keeps decaying for weeks. The fastest complete reversion found anywhere takes about a dozen days.
• And the square-root rule everyone quotes has edges nobody quotes. Below about a thousandth of daily volume it goes linear; a logarithm fits better across a wider range; and one study measured 0.60 rather than 0.50 — for a different quantity from the one usually meant.
• Measured on our own data: across 23,868 structure events in our Structure Lab, the spread cost 0.28R a trade on five-minute charts and 0.03R on daily ones. And the volume histogram under your chart moves almost in step with candle size — rank correlations of 0.71 to 0.80 — which makes it, in practice, a second volatility indicator.
• Your side of it: the regulator's own guidance says a dealer may not claim you have direct access to the interbank market, because it is the counterparty to your trade.
• The closing study is the one that joins the two books. Roughly five million retail trades from about fifteen thousand traders, and a finding that simple strategies built on that aggregated order flow could be profitable — for whoever holds the book.

Tested, not just taught. Besides the Structure Lab, the book's claims about tick volume, volatility clustering, the daily cycle and fixed stops were checked against recorded prices from Bacchus Analytics' own systems. The method and its limits are printed in full.

154 pages. Sixteen chapters, the data section, three appendices — a microstructure card, the numbers and where they come from, and sources — and nine original figures.

An honest note. No firm is accused of anything here. Where one is named it is because it published something, and that published thing is quoted exactly. The mathematics is kept out of the body and put in an appendix with its validity range attached — which the usual retellings leave off, and which is where most of the misuse comes from.

Instant PDF download.

Educational purposes only — not trading, investment, tax or financial advice, and not a recommendation to buy, sell or trade any instrument. Market structure figures reflect published surveys as at their stated dates and should be checked against current releases. Trading leveraged products carries a high level of risk and can result in the loss of some or all of your capital. Bacchus Analytics Limited, trading as Traderz Life.