What a forex trading strategy actually is
A forex trading strategy is a repeatable set of rules that tells you when to enter, when to exit, and how much to risk, and the strategy is the process, not the outcome of any one trade, judged over a large sample. The word gets misused to mean a setup or an indicator signal, but a real strategy is a complete system that produces a distribution of trades over time (ESMA).
The rules have to be specific enough to act on without judgment, because a strategy you can override at will is just a feeling wearing a strategy's clothes. "Buy when the fast average crosses the slow one" is a signal, not a strategy, until it is bolted to a stop, a target, a position size, and a rule for which pairs and sessions.
I treat a strategy as finished only when I could hand it to someone else and they would take the same trades I would, because that test exposes the gaps where discretion was doing the real work. The discipline is in the rules, and the rules are what turn trading from a series of guesses into a process you can measure.
This guide is the hub for the strategies cluster on the site, covering the families, the maths that decides whether they pay, and the honest reasons most retail strategies lose. Read it top to bottom for the framework, or jump to the page you need.
The one number that decides everything
Strip away the indicators and the jargon and a strategy lives or dies on a single number called expectancy, which is the average amount you make or lose per trade over a large sample. A strategy with positive expectancy makes money over enough trades, and a strategy with negative expectancy loses money over enough trades, and no amount of skill in any single trade changes the long-run maths.
The formula is simple and worth memorising. Expectancy equals the win rate multiplied by the average win, minus the loss rate multiplied by the average loss.
A strategy that wins 40 percent of the time, averages 300 dollars on a win and 100 dollars on a loss, has an expectancy of 0.40 times 300 minus 0.60 times 100, which is 60 dollars per trade, and that is a profitable system.
Notice that the 40 percent win rate is fine, because the wins are three times the losses, which is the point most beginners miss. A strategy does not need a high win rate to pay, it needs a positive expectancy, and a high win rate with tiny wins and rare huge losses is the profile of a negative-expectancy system disguised as a winning one.
I calculate expectancy on every strategy before I risk a real dollar, because it is the one number that tells you the truth about a method, and a method that cannot produce a positive expectancy on paper will not produce one in the market either.
The cost floor every strategy must clear
Expectancy has a silent partner that the formula above omits, and that partner is cost. Every trade pays a spread, a round-trip commission, some slippage, and if it is held overnight, a swap, and those four together are a fixed drag that comes off the expectancy before any profit registers.
The drag is what makes the same strategy profitable on one account and losing on another, because the costs are not equal across brokers, sessions, or timeframes. A strategy with a gross expectancy of half a pip per trade is profitable on a 0.2-pip raw spread and dead on a 0.8-pip standard spread, and the strategy's rules have not changed at all.
The shorter the timeframe, the larger the cost slice, because the targets shrink while the per-trade cost stays fixed. On a 3-pip target the spread and commission can take half the gain before slippage, which is why the highest-frequency styles are the hardest to keep in profit, and the honest maths of that is in the guide to tick scalping.
I run every strategy's expectancy net of realistic costs, never gross, because a gross-positive, net-negative strategy is the most common kind there is, and the cost floor is where most retail strategies quietly die.
The four major strategy families
Most strategies are variations on four core ideas, and each idea has a market condition it suits and a condition where it loses money. Understanding the families is more useful than collecting setups, because the family tells you when the strategy should and should not be working.
Trend-following strategies aim to enter in the direction of an established move and ride it, and they suit markets that trend but bleed in ranges where they are whipsawed by false breakouts. Mean-reversion strategies do the opposite, fading moves that have extended too far in the expectation of a return to an average, and they suit ranges but get destroyed in the strong trends they keep trying to fade.
Breakout strategies enter when price escapes a defined range, betting the escape becomes a new move, and they suit markets that coil and release but suffer in the false breakouts that ranges produce. Carry strategies earn the swap on pairs where holding pays a positive overnight rate, and they suit calm, higher-yield environments but suffer when the funding currency suddenly strengthens.
The table sets them side by side, and the contrast matters more than the detail. Each family is a bet on a different kind of market behaviour, which means each family wins when its market shows up and loses when it does not.
| Family | What it bets on | Ideal market | Win-rate profile | Main risk |
|---|---|---|---|---|
| Trend-following | A move continuing | Trending | Low, large wins | Whipsaw in ranges |
| Mean-reversion | A move reverting | Ranging | High, small wins | Runaway trends |
| Breakout | A range escaping | Coiling then releasing | Mixed | False breakouts |
| Carry | Swap income | Calm, higher-yield | High, slow bleed risk | Funding-currency spikes |
I match the family to the market I am actually in, because the most common strategic error is running a trend system through a range or a mean-reversion system through a trend, and the market decides which family is right on any given day.
Timeframe styles: scalping to position trading
The other axis that defines a strategy is the holding period, and the choice of timeframe shapes everything from the cost sensitivity to the lifestyle the strategy demands. The families above can run on any timeframe, but the timeframe changes their character entirely.
Scalping holds trades for seconds to minutes and targets a few pips, which makes it the most cost-sensitive style and the most demanding of a raw-spread account and fast execution. Day trading closes every position by the end of the session, which removes overnight risk and swap cost but still demands constant attention through the trading day.
Swing trading holds for days to weeks and targets larger moves, which makes costs a rounding error and lets the edge breathe, and it suits people with day jobs. Position trading holds for weeks to months and trades the big picture, which demands patience and the temperament to sit through large unrealised swings.
I chose my timeframe to fit my life before I chose my strategy, because a strategy you cannot actually sit and trade is a strategy you will abandon at the worst moment. The honest comparison of the fastest styles is in the guide to the 20 pips a day challenge, and the day-trading end is covered in the 4 hour time frame strategy.
Win rate versus risk to reward
The two halves of expectancy, win rate and risk to reward, are locked in an inverse relationship, and understanding the trade-off is the key to picking a strategy you can actually execute. You can win often with small gains relative to your losses, or win rarely with large gains relative to your losses, and both can be profitable if the expectancy is positive.
A trend-following system typically has a modest win rate, often 35 to 45 percent, with average wins two to four times the average losses, and it pays because the occasional big trend covers the many small false starts. A mean-reversion system typically has a high win rate, often 65 to 80 percent, with many small wins and occasional large losses, and it pays as long as those large losses stay controlled.
The danger lives at the extremes of each profile. The high-win-rate system lulls you into sizing up because the wins feel easy, and then one outsized loss wipes months of gains, while the low-win-rate system tests your discipline with long losing streaks that feel unbearable even when the maths is sound.
I pick the profile I can stick with through its bad streak, because the strategy only pays if you trade it through the drawdown, and a positive-expectancy system you abandon in its losing run pays exactly nothing.
Backtesting, and why most winners lose live
Backtesting is how you estimate a strategy's expectancy before you risk money, and it is also the step where most retail strategies pick up a false positive. A backtest that shows a clean equity curve often reflects the tester's choices more than the strategy's edge, and understanding why is the difference between a plan and a fantasy.
The first trap is overfitting, where the rules are tuned and retuned on the same historical data until they fit the past perfectly, which guarantees they fit nothing else. A strategy with twenty optimised parameters that nails the last five years is a strategy that has memorised the past rather than learned from it, and it will fail on data it has not seen.
The second trap is omitting costs, where the backtest counts gross pips and ignores the spread, commission and slippage that real trading pays. A strategy that wins a fraction of a pip per trade gross is a cost-negative strategy in disguise, and the test that does not subtract costs per trade is hiding the most important number.
The third trap is survivorship and selection, where the tester discards the losing variations and reports only the winner, which is the same error as the social-media trader who posts only the green days. I run backtests with realistic costs, out-of-sample data the rules were not tuned on, and the full set of variations including the losers, because a backtest that survives those three filters is the only kind worth acting on.
Why most retail strategies lose
The loss rate among retail forex traders is high and remarkably consistent, and the honest reasons are structural rather than personal. ESMA's data shows 74% to 89% of retail accounts lose money, and the figure is stable across brokers and regions because the causes are stable (ESMA).
The first cause is negative expectancy after costs, which is the cost floor above applied to strategies that looked profitable gross. The second is overfit backtesting, which is the false positive that makes a losing strategy look like a winner until real money tests it.
The third cause is behaviour, because real trading with real money produces fear, greed and revenge that no backtest models, and these emotions convert a positive-expectancy system into a negative-expectancy result. The fourth is leverage, because the strategies most beginners run depend on the small-account, high-leverage combination where the loss data concentrates, and the leverage turns ordinary adverse moves into account-enders.
I do not say no retail trader wins, because some do, but the winners share a pattern: a tested net-of-cost edge, a risk rule they follow under stress, and a timeframe and leverage that let the edge survive. The losers share the opposite pattern, and the gap between them is process, not luck.
How to build a strategy that survives
The honest way to build a strategy is a sequence that filters out the false positives before they reach your wallet, and skipping any step is how good backtests become blown accounts. The process is dull on purpose, because the dullness is what keeps you out of the strategies that only look good.
Start with a hypothesis, a clear idea of why a particular rule should produce an edge, because a strategy with no reason behind it is a pattern you are trusting on faith. Test the hypothesis on historical data with realistic costs and out-of-sample validation, and if the net expectancy is not clearly positive, the hypothesis fails and you move on.
Forward-test the survivor on a demo account for weeks, because demo reveals whether you can actually execute the rules in real time, and many strategies that pass a backtest fail a human. Then run it on a small live account, because live money reveals whether you can execute the rules under emotion, and that is the test that matters.
I review the live results against the backtest expectancy regularly, because a strategy whose live results diverge from its tested expectancy has either degraded or was never real, and either way it is time to stop. The full risk framework that keeps a surviving strategy survivable is in the guide to volatility-based position sizing.
The guides in this cluster
Each guide below is self-contained, and I have grouped them by the question they answer so you can read the one that matches where you are. The two demand pages take apart viral strategy claims honestly, and the day-trading pages cover the practical end of the timeframe spectrum.
Indicators versus price action
Strategies are built on either indicators, price action, or a blend, and the choice is less important than the honesty about what each can and cannot do. Indicators are maths applied to past price, and they summarise what already happened in a form a rule can act on.
The strength of indicators is that they are unambiguous, because a moving average cross either happened or it did not, which removes the judgment that erodes discipline. The weakness is that they lag, because they are derived from past data, which means they confirm moves after they have begun and signal reversals after the turn.
Price action reads the raw candlesticks and structure, and its strength is that it is timely, because price is the most current information there is. Its weakness is that it is subjective, because two traders can read the same chart differently, which puts the discretion, and therefore the inconsistency, back into the strategy.
I use indicators to enforce rules and price action to read context, because combining the two plays to each one's strength, and the foundational price-action reading is in the guide to support and resistance.
Session and pair selection
A strategy's results depend heavily on which pairs and sessions it trades, because liquidity and volatility vary across both, and the same rules produce different expectancy in different conditions. The London and New York overlap is the most liquid window, which means the tightest spreads and the most reliable moves for most strategies.
The Asian session is quieter and more range-bound, which suits mean-reversion strategies but starves trend and breakout systems. The pairs matter too, because the majors offer the tightest spreads and cleanest behaviour, while exotic pairs offer larger moves at the cost of wider spreads that often erase the edge.
I define the session and pair list as part of the strategy, not as an afterthought, because a strategy tested on EURUSD during London that is then traded on an exotic pair during Asia is a different strategy with a different expectancy. The rules and the conditions are the strategy together.
Trading the wrong session is one of the quiet profitability killers, and the pair-and-session choices are part of why the same setup pays for one trader and loses for another.
Risk management as part of the strategy
A strategy with a positive expectancy can still blow up an account if the position sizing is wrong, which makes risk management part of the strategy rather than a separate concern. The expectancy tells you the per-trade average, and the sizing tells you whether you survive long enough for the average to play out.
The rule that keeps a positive-expectancy system alive is fixed fractional risk, meaning you risk a small, constant percentage of the account on each trade, typically half a percent to one percent. This makes the lot size a function of your stop distance and your risk percent, so the position shrinks as the account shrinks and you never face the ruin that fixed-lot sizing invites.
The danger of ignoring sizing is that a string of losses, which every strategy produces, becomes unrecoverable when the lots stay fixed while the balance falls. A 50 percent drawdown needs a 100 percent gain to recover, which is the maths that turns a normal losing streak into a terminal one under fixed-lot sizing.
I treat the risk rule as the most important part of the strategy, because a mediocre strategy with great sizing survives, and a great strategy with bad sizing does not, and the method is in the guide to volatility-based position sizing.
The honest expectation
I want to end the framework with the honest expectation, because the marketing around forex strategies sets expectations that guarantee disappointment. A skilled retail trader making 5 to 10 percent a month is doing very well, and anything promising doubling in a week or a month is selling something the market does not offer.
The realistic path is a tested edge, executed with discipline, compounding slowly over months and years, with drawdowns along the way that test whether you actually trust the system. The strategies that survive this path are the boring ones, and the strategies that go viral are the ones that do not survive it, which is the selection bias behind most strategy content.
The wider context on whether the whole enterprise can pay is in the guide to whether forex trading is profitable, and the structured self-study to build the foundation is the free forex course. A strategy is the top of a pyramid whose base is risk, costs and behaviour, and the base is what decides whether the top stands.
Common strategy mistakes
The mistakes that sink strategies are repetitive, and naming them is most of avoiding them. The first is conflating a setup with a strategy, taking an indicator signal with no stop, target or size, which is a guess rather than a system.
The second is overfitting the backtest, tuning the rules until the equity curve is perfect on past data, which guarantees the rules fail on future data. The third is omitting costs from the test, producing a gross-positive, net-negative strategy that loses from the first live trade.
The fourth is abandoning the system in its drawdown, which is the most expensive mistake because it denies you the expectancy you validated. The fifth is sizing by confidence, scaling up on the trades that feel certain, which means the wrong calls cost the most.
I keep the defence simple: a complete strategy with a tested net expectancy, a fixed risk rule, and the discipline to trade it through the bad streak, and most of the mistakes above fall away at one of those gates. The strategies cluster on this site exists to take those gates seriously, one page at a time.