What Is a Trading Strategy? A Complete Beginner Guide

The plain-English definition

A trading strategy is a written set of rules that decides, without asking your opinion on the day, which asset you buy or sell, at what moment, how much of it, and under what conditions you close the position. It is closer to a recipe than a forecast. The recipe does not care whether you feel confident; it simply lists the steps.

The difference matters because markets punish improvisation. Prices move for reasons that no individual can fully observe, so any single trade is close to a coin flip with a small tilt. A strategy is the machinery that converts that small tilt into a result over hundreds of trades, in the same way a casino converts a 5 percent house edge into reliable revenue across millions of hands while losing plenty of individual ones.

What a strategy is not

• Not a prediction. You are not forecasting where the S&P 500 will be in December. You are defining a response to conditions as they appear.
• Not an indicator. The relative strength index is a calculation, not a plan. An indicator becomes part of a strategy only when you attach a specific action to a specific value of it.
• Not a tip or a signal service. A signal tells you to buy now. A strategy tells you why, how much, when to exit, and what to do when it fails.
• Not a guarantee. Every real strategy loses money for stretches of time. Strategies that never lose in a backtest are almost always fitted to the past rather than describing the market.

The six questions every strategy must answer

Professional traders describe a strategy in terms of its decision points. If any of the following six is missing, you do not yet have a strategy, you have half of one, and the missing half is where the losses accumulate.

1. 1

   What do you trade, and on what timeframe?

   Name the universe precisely: "US-listed stocks in the S&P 500 with average daily volume above 2 million shares", or "EUR/USD spot", or "front-month WTI crude futures". Then name the bar size you make decisions on: 1-minute, hourly, daily, or weekly. The same rules produce completely different results on different timeframes.

2. 2

   What triggers an entry?

   The condition that must be true for you to open a position, written as something a computer could check. For example: "the closing price is above the highest close of the prior 50 sessions and the 200-day moving average is rising".

3. 3

   How do you exit a winning trade?

   A target price, a trailing stop, a time limit, or a signal reversal. Without this rule, profitable trades quietly turn into losing ones. Many beginners define entries in obsessive detail and leave exits to instinct, which is backwards: exits determine the size of your average win.

4. 4

   How do you exit a losing trade?

   The price or condition that proves the idea wrong. This is the stop loss. It is defined before entry, based on where the market structure invalidates your reason for the trade, never on the amount of money you are willing to see in the red.

5. 5

   How large is the position?

   Position size converts a signal into risk. The standard approach: risk a fixed small fraction of account equity per trade, commonly 0.25 to 1 percent, and calculate the share or contract count from the distance to your stop. See position sizing for the arithmetic.

6. 6

   When do you stop trading the strategy?

   Every strategy eventually degrades as markets change. Decide in advance what evidence would retire it: a drawdown deeper than anything in your backtest, a run of losses statistically inconsistent with the tested distribution, or a structural change such as an exchange rule change or the disappearance of the inefficiency.

Why any strategy can make money at all

If markets were perfectly efficient, no rule would beat buying and holding. They are not perfectly efficient, but the inefficiencies are small, crowded, and temporary. Almost every durable strategy earns its return from one of four sources, and knowing which one you are harvesting tells you when the strategy will stop working.

Notice what is absent from this table: "the chart made a pattern". Patterns matter only when they are a visible fingerprint of one of these four forces. A breakout works, when it works, because traders who missed the move chase it and because stop orders cluster above obvious highs, not because the shape itself has power.

Expectancy: the only number that decides if a strategy is worth trading

Beginners obsess over win rate. Win rate alone is meaningless. A strategy that wins 90 percent of the time and loses ten times its average win on the remaining 10 percent is a slow-motion disaster. The number that matters is expectancy: the average profit or loss per trade once wins and losses are weighed by how often they occur and how large they are.

Expectancy = (Win rate x Average win) - (Loss rate x Average loss)

Example A: trend following
  Win rate        35%      Average win    3.0R
  Loss rate       65%      Average loss   1.0R
  Expectancy = (0.35 x 3.0) - (0.65 x 1.0) = 1.05 - 0.65 = +0.40R per trade

Example B: high win rate, poor payoff
  Win rate        85%      Average win    0.5R
  Loss rate       15%      Average loss   4.0R
  Expectancy = (0.85 x 0.5) - (0.15 x 4.0) = 0.425 - 0.60 = -0.175R per trade

The letter R stands for the amount you risk on one trade, so a 3R win means you made three times what you were prepared to lose. Using R instead of dollars lets you compare strategies across account sizes and markets. Example A wins barely a third of the time and is excellent. Example B wins almost every trade and destroys the account. Trend following feels terrible to trade and makes money; selling insurance against disasters feels wonderful and occasionally ends the game.

A complete strategy, written out in full

Below is a deliberately simple trend-following strategy, written at the level of precision that makes it testable. It is an illustration of format and completeness, not a recommendation, and its parameters are the obvious ones rather than optimised ones.

Universe

The 100 largest US-listed stocks by market capitalisation, rebalanced quarterly, excluding any stock with average 20-day dollar volume below 20 million USD.

Timeframe

Daily bars. All decisions are made on the closing price and executed at the next open.

Entry

Buy at the next open when today’s close is the highest close of the last 100 trading days AND the 200-day simple moving average is higher than it was 20 days ago.

Initial stop

Place a stop at the entry price minus 2.5 times the 20-day Average True Range. This adapts the stop to each stock’s own volatility.

Trailing exit

Once in profit, raise the stop each day to the higher of the previous stop and the close minus 2.5 ATR. Never lower it. Exit on the open following a stop breach.

Position size

Risk 0.5 percent of account equity per position: shares = (0.005 x equity) / (2.5 x ATR). Cap any single position at 10 percent of equity.

Portfolio limits

Maximum 10 open positions. No new entries if total open risk exceeds 4 percent of equity. Maximum 3 positions in any one sector.

Retirement condition

Stop trading and re-examine if the live drawdown exceeds 1.5 times the worst drawdown observed in the 20-year backtest, or if 12 months pass with returns below the worst rolling 12-month backtest result.

Every line is checkable by a computer, which is precisely the point. Whether this particular strategy is profitable is a question for a backtest and then for out-of-sample validation, but it is at least a real hypothesis that can be proven wrong.

Worked example: from signal to trade to outcome

Assume a 25,000 USD account trading the strategy above. On a Tuesday close, stock XYZ prints its highest close in 100 days at 84.20 USD, and the 200-day average is rising. The 20-day ATR is 2.10 USD.

Three outcomes are possible, and the strategy has a defined answer for each. If the stock falls to 78.95 in the next week, you lose roughly 125 USD, which is 0.5 percent of the account, and the loss is unremarkable. If the stock trends to 100 USD over three months with the trailing stop dragged up to 94.50, you exit with about 10.30 USD per share, roughly 237 USD, which is 1.9R. If the stock chops sideways for two months and eventually stops you out at a slightly higher trailing stop, you take a small gain or a small loss and free the capital.

The single trade tells you nothing about whether the strategy works. Only the distribution of a few hundred such trades does. This is the mental shift that separates a trader with a strategy from a gambler with a conviction.

How to build your first strategy, step by step

1. 1

   Start with an observation, not an indicator

   Something you believe is true about market behaviour: "stocks that fall hard on no news tend to bounce", or "currencies trend for months once central bank policy diverges". The observation is your hypothesis about which of the four return sources you are harvesting.

2. 2

   Translate it into measurable conditions

   "Falls hard" becomes "closes 3 or more ATRs below the 10-day average". "No news" becomes "no earnings report within 5 sessions". Each translation is a decision you will later test.

3. 3

   Define the exit before the entry

   Decide what invalidates the idea and what caps the reward. Attach a time stop too: if the expected behaviour has not occurred within N bars, the reason for the trade has expired.

4. 4

   Add position sizing and portfolio limits

   Fixed fractional risk per trade, a cap on simultaneous positions, and a cap on correlated exposure. Without these, one bad week can erase a year of edge.

5. 5

   Backtest on a long history, honestly

   Include costs, use point-in-time data, and never let a rule see information that was not available at decision time. Read look-ahead bias before you trust any result.

6. 6

   Validate out of sample and forward test

   Hold back the most recent few years, then run the strategy in paper trading for long enough to see a realistic number of trades. Expect live results to be meaningfully worse than the backtest.

7. 7

   Start small and scale slowly

   Trade minimum size for a quarter. The goal is to confirm that your execution, data, and discipline survive contact with a real account, not to make money quickly.

Mistakes that quietly destroy beginner strategies

• Optimising until the equity curve is beautiful. Every extra parameter fits the past better and the future worse. See overfitting.
• Changing the rules mid-trade. Moving a stop lower "just this once" converts a defined 1R loss into an undefined one. Most account-ending losses start here.
• Trading a strategy whose drawdown you cannot tolerate. If the backtest shows a 30 percent drawdown lasting 14 months, assume you will experience worse, and that you will want to quit at the bottom.
• Ignoring costs. Spread and slippage kill most short-term strategies that look profitable on closing prices.
• Running several strategies that are secretly the same trade. Five momentum strategies across correlated stocks is one large position wearing five hats. See correlation and diversification.
• Judging a strategy by its last five trades. With a 40 percent win rate, five consecutive losses happen roughly every thirteen trade sequences. That is noise, not evidence.

Where to go next

If markets themselves are still unfamiliar, read how financial markets work and how to read a price chart first. If you are ready to pick a style, the strategy types overview compares every major family side by side, and best strategies by asset class matches those families to the market you actually trade.

Whatever you choose, the sequence is always the same: write the rules, test them on history, test them on data the rules never saw, size positions so that being wrong is boring, and keep a record. Everything else in this library is detail on top of that skeleton.