The Achilles Trading System

Welcome

What the Achilles Trading System is, and what it is for

The Achilles Trading System is a family of five indicators built to work as one. They were made because most trading tools share the same problem: they tell you what to do, and they never tell you how often that worked. These do the opposite. They are built to describe the market accurately — where the important prices are, how stretched things have become, whether real money is moving — and then hand the decision back to you, with an honest note about how much each reading is actually worth.

The five pieces cover different jobs and are designed to be read together:

• Achilles A puts the reference prices on your chart — the bands, the key averages, the day's volume-weighted average price — and keeps a running confluence score in the corner.
• Achilles B is the momentum pane: one clear wave, a handful of readings, and a compact card that tells you the state of everything at a glance.
• Achilles B Classic is the same engine in the denser, more familiar layout, for traders who already think in that visual language.
• Achilles Bar adds context strips — how volatile, how heavy, how bought or sold this session is.
• Achilles SMC does the structural work: it finds the zones, the untaken highs and lows, and the support and resistance that everything else is measured against.

A word on how to read this book. Every feature carries a small label saying whether it ships switched on, whether it is still experimental, and — where we have tested it — what the test found. Some of those labels say the feature does not predict anything. Those are the honest ones, and they are there on purpose: a tool that describes the market well is genuinely useful, and knowing which of your tools are description and which are prediction is most of what separates a consistent trader from a hopeful one.

Start here

Three things to know before you turn anything on

The suite is deliberately honest about what it knows. Most indicators tell you they predict things. These have been measured, and most of them don't — they describe. That is still useful, and it is a great deal more useful than a signal you trust for no reason. Every feature in this book carries a label:

The five indicators, in one table

Achilles A

Sits on the price chart · price levels and confluence

What it draws

The Confluence Scorecard — the main event

This is the part of Achilles A worth learning properly, because it is the one reading in the whole suite with evidence behind it. Read the two percentages, not the ticks.

• Both low and close together — as in the 32% / 31% above — means the checks disagree with each other. That is a market with no consensus, and it is the most common state.
• One side clearly ahead means many unrelated measurements are pointing the same way at once. That agreement is the signal, not any individual row.
• A single tick means nothing. The card works because fifteen things agree, and any one of them alone is noise.

The weights are not opinions. The daily session VWAP carries 14 because it held up in testing; money flow carries 5 because it did not. That is the difference between a scorecard and a wishlist.

The markers on your candles

Achilles A prints small shapes above the candles. Each one means something specific, and there are only five to learn:

How to use it

Read the SR VWAP first. It answers one question cleanly — is price above or below today's average cost? Of the eleven components we measured individually, it was one of only three that held up on both halves of the test, and it carries the heaviest weight on the scorecard.

Then read the scorecard total, never a single row. The card works through breadth — the fact that many unrelated things agree. One tick on one row means nothing at all.

Achilles B

Own pane · the next-generation wave design

What it draws

How to read the card

The card is the fastest way in. Each line is one plain-English state rather than a number you have to interpret:

Achilles B Classic

Own pane · the original design

Classic adds, on top of everything in Achilles B:

Achilles Bar

Own pane · made to sit with Achilles B

The Heel strip ships off here for a specific reason: Achilles B already shows the Heel. The same reading appearing twice on one screen looks like two pieces of evidence when it is one. Turn it on if you are running Bar without Achilles B.

Achilles SMC

Sits on the price chart · structure, zones and levels

What SMC gives you that a blank chart does not

Most traders mark structure by hand: draw a line at the high, another at the low, shade a zone where price took off, redraw it all when the timeframe changes. SMC does that work continuously and consistently, and it does it the same way every time — which is the part hand-drawing cannot promise.

How to use it

Blocks and magnets are the two to learn first. A block tells you where price previously left in a hurry. A magnet tells you where unfilled orders probably rest. Between them you get a short list of prices worth caring about — which is exactly the "structure first" the rest of the suite is built on.

What ships on, what ships off

The exact settings a new install starts with

Add all five indicators to a fresh chart and this is what you get. The principle behind the list is simple: a first chart should be readable, not complete. Everything switched off is still there, one tick away, waiting for you to have a reason to want it.

Achilles A — everything on

All five features run by default: the Achilles Bands, The Achilles Line, the 200 EMA, the SR VWAP, the Whale Detector flags and the Confluence Scorecard. Nothing here is noisy enough to need hiding, and the scorecard is the reading we most want you to see.

Achilles B — everything on

The wave and its trigger, the buy and sell dots including the live one on the forming bar, money flow and its pressure line, the Momentum Line, RSI, the Achilles Heel, and the full readout card with the SR VWAP row and the Confluence score. The glow and the gradient wash are on too — they are what make one wave readable where four lines would not be.

Achilles B Classic — everything on except one

Achilles Bar — three of four strips

Achilles SMC — the one with real choices made

This is where the defaults do the most work, because SMC can draw far more than any chart can carry.

How this compares to other momentum indicators

An honest answer, including where we are the same

If you have used a momentum-style oscillator before, the wave in Achilles B will look familiar — a line that rises and falls around a centre, with marks where it turns. That is deliberate. Momentum is a well-understood idea and there is no advantage in reinventing how it is drawn, so the pitch here is not that we invented a better wave.

We ran the comparison ourselves

The obvious question about any of these tools is whether the money-flow reading actually predicts anything. So we tested it — ours and theirs, on the same bars:

Neither one predicts direction. That is not the answer we hoped for, and it is the answer. Our test ran across three coins and four timeframes with a proper holdout, and an early result that looked promising on one coin over two months reversed sign once the full sample was in — which is exactly why single-chart impressions are worth so little.

So money flow is on the chart because it describes buying and selling pressure, and it carries a weight of 5 out of 100 in our Confluence Scorecard. That number is not a guess. It is what the measurement earned it.

What you get here that we have not seen elsewhere

What we actually tested

Every result, in one table

Most indicator documentation has no section like this. This one exists because the alternative — asking you to trust a tool because it looks like it works — is how people lose money slowly.

The headline: nothing in this suite has been shown to predict direction. Not one feature, including the one we built the scorecard around. What follows is the evidence, and then why these tools are still worth having on your chart.

How to read the results

• Measured null — the test could have found a useful effect and did not.
• Could not tell — the effect, if any, was smaller than the test could resolve. Not the same as "no effect".
• Withdrawn — a result we published and later found to be an artefact of how it was measured.
• Description — verified to draw what it claims to draw; never tested as a forecast.

Samples count non-overlapping events, not bars. An effect must exceed 2.8 times its own standard error to count. For scale: the largest effect this project has ever measured is +0.024R — about 2.5% of one unit of risk. Treat any larger claim, from anyone, with suspicion.

The results

Read the control row. Across 2,494 samples a bar's own direction continues essentially never — the average outcome is zero. That is the bar every result above has to clear, and none of them do.

The one we got wrong

The Confluence Scorecard measured +2.7 percentage points on unseen data, and for a while we believed it. The control was wrong: a signal that trades both ways was being scored against an always-long benchmark on a sample that drifted down, so it collected points for its direction mix rather than its timing. Scored against its own mix, the edge fell to +0.31 and +0.37 points across two holdouts — zero.

We found it, we corrected it, and we withdrew our only positive result. That is what the process is for.

Two things that are arithmetic, not prediction

• The fee floor is 0.20% of price. Below that stop distance, costs eat more than 15% of your risk — about $4.80 at a $2,400 price, and short-timeframe stops routinely fall under it.
• A tighter stop costs more, not less. Halving your stop distance doubles what fees take as a share of your risk. This is the most commonly inverted idea in retail trading.

Neither predicts anything. Both change which trades are worth taking, which in practice matters more than most signals would even if they worked.

What this means for you

Every indicator on a chart is doing one of two jobs: describing what has happened, or predicting what will. Description is reliable — where price left in a hurry, where orders rest, how stretched things are, whether this session is busy. Prediction is where nearly every claim in this industry, including ours, falls apart under measurement.

These tools are good at description. Use them to build a short list of prices worth caring about, and to know what kind of market you are in when price arrives at one. Then make the decision yourself, with a stop wide enough that fees do not eat it.