Independent provider directory
The method Strategy playbook How-to guides FAQ See the tested method
The method

How a forex trading strategy actually works

A forex trade is a wager on the relative price of two currencies over the next stretch of time. Turning that wager into a strategy means writing down exactly which pair you trade, when, on what trigger, and how you get out — before you are in the position and tempted to manage it by mood.

The mechanics, in order

The ground truth: pairs, sessions and leverage

Forex prices are always a ratio. A currency pair tells you how much of one currency it takes to buy the other, so you are never long a single thing — you are long one currency and short the other at the same time. The market trades around the clock in overlapping sessions, and the hours you keep shape which strategies even make sense: a quiet session rewards patient range plans, an active overlap suits trend and breakout plans. And it runs on leverage, which magnifies both the move you caught and the one you got wrong. Every honest forex strategy is built with that magnification in mind first, not last.

The three numbers fixed before you enter

A feeling says “this pair looks ready.” A strategy says three things, in writing, before the position opens:

  • Entry — the exact level or condition your rules act on, not “somewhere around here.”
  • Stop — the level that proves the idea wrong, chosen before you are emotionally invested in it.
  • Target — where the move you expected is judged complete, so profit is taken on plan rather than on nerve.

Those three are not three separate decisions. They are one decision, made together, before the entry exists. The schematic below shows where each sits in the one situation this whole site keeps returning to: a pair that has stretched a measured distance from the middle of its own recent range.

Where a mean-reversion fade sets its entry, stop and targetSchematic of a currency pair that has stretched a measured distance above the middle of its own recent range. A mean-reversion rule fires a short at the upper edge (the entry), places the stop a fixed distance further out where a continued push would prove the stretch was a new move instead, and sets the target back near the middle where the expected snap-back is judged complete.PRICE, stretched from the middle of its own recent range →STOPENTRY (fade)TARGET / middlerule fires herethe stretch buildsreversion toward middle
Illustrative schematic, not a specific recommendation. The three levels — entry at the stretched edge, stop beyond it, target back near the middle — are all fixed before the position opens, which is what makes the trade a clean test of a rule rather than a feel.

A worked setup, start to finish

Reading the levels off a diagram is one thing; watching the logic run is another. Here is a single illustrative setup carried from trigger to outcome — a made-up example for teaching, not a specific recommendation to trade any pair.

Worked setup · illustrative, not a recommendation
  1. The situation. A pair has pushed up for several sessions and now sits a defined distance above the middle of its recent range — further than it has typically strayed before turning. Say it trades at 1.0940, with the middle of its range near 1.0890.
  2. Why this entry. The rule, not the trader's opinion, treats that stretch as the signal. The plan fires a short at the stretched edge: entry 1.0940. You are not predicting a top; you are acting on a measured distance the rule defined in advance.
  3. Where the stop goes. The stop sits a fixed distance further out, at 1.0975 — 35 pips beyond the entry. If price pushes there, the “stretch” was not a stretch at all but the start of a genuine new move, and the idea is admitted wrong before it costs more.
  4. What the target is. The target sits back near the middle the price reverted from: 1.0895, about 45 pips of expected snap-back. That is roughly 1.3 units of reward for every 1 unit risked — modest, repeatable, and decided before the entry.
  5. What invalidates it. Two things end the trade on plan: price tagging 1.0975 (stop, a small pre-decided loss) or 1.0895 (target, the planned win). A third — the holding window closing with neither hit — closes it flat-ish either way. Nothing is decided after the position is open.
  6. The realistic outcome distribution. A plan like this does not win every time, and it is not supposed to. Across many such trades you would expect a majority to revert and reach the target, a minority to push through the stop, and the edge to live in the gap between a slightly-better-than-even hit rate and a reward that modestly outweighs the risk. One trade tells you almost nothing; the distribution of a hundred tells you whether the rule has an edge.

Sizing the trade: the arithmetic

The setup above says where. Position sizing says how much — and on leverage it is the decision that keeps you solvent. It is pure arithmetic, worked here on different numbers from any other page so you can see the method rather than memorise a figure.

Worked example · illustrative, not a recommendation

The arithmetic that turns a risk rule into a position size never changes; only the numbers do. Suppose a hypothetical account and a single rule: never risk more than a fixed slice of it on one trade.

  1. Account and per-trade cap. Account is $8,000; the rule caps risk at 1.0% of it per trade. That is $80 you are willing to lose if the stop is hit — decided before the entry, not after.
  2. Stop distance. The plan's stop sits 35 pips from the entry — the level beyond which the idea is wrong. At an illustrative $1.00 per pip per unit, each unit of size risks 35 × $1.00 = $35.00 if stopped.
  3. Position size falls out of the cap. Divide the cash you may risk by the risk per unit: $80 ÷ $35.00 ≈ 2 units. The size is whatever keeps the loss at the cap — it is an output of the stop and the rule, never a number you reach for because the setup “feels” strong.
  4. Widen the stop, shrink the size. Double the stop distance and the same cap buys you roughly half the units. The cap is the constant; the size bends to honour it. That single habit is what lets a strategy survive a losing streak long enough to be judged.

Where conviction comes in: a grade does not change the cap, it tells you where to lean inside it. If a method grades its calls A to D, you might size an A toward the top of your cap and a D toward the bottom — an A still risks no more than the rule allows, it just uses more of the room the rule gives you. Grading is a dial within the cap, never a licence to breach it.

What separates a strategy from a hunch

The dividing line is testability. A real strategy can be run against history and forward in time and produce a record — a count of trades, a win rate with the losses left in, a worst drawdown. A hunch produces only screenshots of the trades that worked. The cleanest proof that a strategy was a strategy and not a story is that each call was written down before its outcome was known.

What a bad version of this looks like

Teaching is only honest if it shows the fragile version too. The same setup goes wrong in predictable ways:

  • No stop, “I'll watch it.” A fade with no hard stop is not a range trade; it is an open-ended short into a pair that might be breaking out. On leverage, “watching it” is how a 35-pip plan becomes a 200-pip loss.
  • Moving the entry to feel earlier. Firing at 1.0925 because the pair “looks toppy” rather than waiting for the rule's level turns a tested edge into a guess, and a guess has no record.
  • Sizing by conviction instead of by stop. Doubling the size because the setup feels strong breaks the cap the arithmetic above exists to protect. Conviction leans inside the cap; it never replaces it.
  • Counting only the winners. Remembering the reverts and forgetting the breakouts makes any fade look brilliant — right up until the unrecorded losses arrive in your account balance.

How do you know a strategy actually works?

Eventually every strategy faces the only question that matters: is the record real, or is it a story with a chart attached? The honest answer is a record you can re-check. A method that publishes its calls, anchors each one to a public ledger at the moment it is sent, and grades each A to D before the outcome is known has turned “trust me” into “check it.” Here is what that chain looks like end to end.

How a published call becomes a claim a stranger can re-checkFlow diagram in four stages: a currency-strategy call is published with its entry, target, stop and A-to-D grade; those fields are turned into a single cryptographic fingerprint; the fingerprint is anchored to a public ledger at the moment of publication, before the trade resolves; later, anyone re-runs the fingerprint on the published call and confirms it matches the on-chain receipt, proving the call was fixed in advance.PUBLICATION TIME → (the receipt is dated before the trade can resolve)A match proves the call existed in exactly this form before the outcome was known.1 . WRITEentry / targetstop / grade+ signal time2 . FINGERPRINTone cryptographicfingerprint ofthose fields3 . ANCHORwritten to a publicledger at publication,before it resolves4 . RE-CHECKanyone re-runs it+ confirms thepublic receipt
This is the gap between a strategy you can cheer and one you can take apart line by line: the record is frozen in public at publication, so no level can be quietly edited after the market moves.

What a conviction grade means — and the bar per clock

A grade is only useful if it is calculated, not chosen. On the systematic method this site points to, each call carries a letter from A to D set against where it sits in that model's own measured return distribution. Because the bar is per clock, an A means the same thing — “top-band for this holding time” — whether the position is carried for an hour or a month:

The grade-A bar is the average per-trade return that earns the top conviction letter on each model — set against that model's own clock, not one figure stretched across all of them.
ModelHolding clockGrade-A bar (avg per trade)
Swing Tradecarried roughly 7 to 28 days6.00% avg / trade
Multi Hourclosed within half a session to two sessions4.50% avg / trade
Day Tradeopened and closed in the same session, inside a 0 to 60 minute window0.70% avg / trade
Investingcarried over a long horizonlong-horizon

An A is the top band of a model's own measured return distribution; D is the lowest letter still published. Because the bar is set per clock, an A on a same-session call (around 0.70% a trade) and an A on a multi-week swing call (around 6.00%) both read as “top-band for this holding time” rather than one absolute target stretched across very different horizons. There is no E grade — it was retired from the live product so the four-step scale keeps its meaning.

From a tested method to a forex track record

The method this site recommends, the #1-ranked provider, writes every call onto the Bitcoin ledger the moment it is sent, which makes “decided in advance” something a stranger can verify rather than take on faith. The forex evidence that the same discipline works on currencies is kept separately and externally: an organiser tracked a 168% return for 4th place in the Annual Forex division of the 2025 World Cup Trading Championships, part of a 294% aggregate across the divisions entered. We keep the two clearly apart — the competition result is the documented forex skill, and the four mean-reversion models are described only by their own published record, by holding clock and count, never by the instruments behind them. To run the verification yourself, follow how to verify a forex strategy is honest.

Related