Risk management in trading is a system of layered limits: a fixed percentage risked per trade, a written invalidation for every position, a position size derived from those two, a daily loss cap, and a drawdown rule that cuts exposure before the account is in trouble. Each layer catches what the one above it misses.
Ask most traders what risk management is and you get one answer: a stop loss. That is one component of one layer, and on its own it does almost nothing. A trader with a perfect stop on every trade can still lose a quarter of an account in an afternoon by taking twelve of them at four times normal size. The stop was never the problem.
What follows is the whole system, in the order the layers should be built. It is the pillar for everything else in this cluster — each layer links to the page that covers it in depth.
Why this is the part that decides the outcome
The uncomfortable thing about retail trading results is how consistent they are across markets and decades. When the UK's Financial Conduct Authority reviewed retail contract-for-difference providers in 2015, its finding was blunt: "According to our sample data, 82% of clients lost money on these products" (FCA, CP16/40). The regulator's response was not to teach better chart reading. It was to cap leverage, standardise risk warnings and force firms to publish their own loss ratios — every one of those a constraint on size and exposure, not on analysis.
That is the shape of the problem. Retail traders do not primarily lose because they cannot read a chart. They lose because nothing caps what a bad idea, a bad day or a bad month is allowed to cost.
The five layers
| Layer | What it controls | What it catches |
|---|---|---|
| 1 · Risk per trade | Cost of being wrong once | Oversizing on conviction |
| 2 · Invalidation | Where the idea is dead | Trades held past the thesis |
| 3 · Position sizing | How layers 1 and 2 connect | Size chosen by feel |
| 4 · Daily & weekly caps | Cost of a bad session | Revenge trading, tilt |
| 5 · Drawdown rule | Cost of a bad month | Grinding an account to nothing |
Each layer exists because the one above it has a specific blind spot. Risk per trade says nothing about how many trades you take, so you need layer 4. Layer 4 says nothing about a slow bleed across six weeks, so you need layer 5.
Layer 1 — risk per trade
Fix, in advance, the percentage of current account equity you are willing to lose on a single position. Commonly 0.5% to 2%. Percentage matters rather than a cash figure, because a percentage shrinks your size automatically during a drawdown and lets it grow as the account recovers — the sizing adapts without you having to make a decision while you are losing.
The survival arithmetic is what makes this the first layer. At 1% risk, twenty consecutive losses still leaves roughly 82% of the account; at 10% risk the same streak leaves about 12%. The full table, and when 1% is the wrong number in either direction, is in the 1% rule in trading.
Layer 2 — a written invalidation
Before entry, write the specific condition that would prove the idea wrong. Not "if it looks bad" — a price, a close, a failure to hold, or a time by which the move should have happened.
This layer is the one most often skipped, and skipping it disables everything below. You cannot size a position without knowing the distance to invalidation, and you cannot enforce a daily cap when every loss is negotiable. The distinction between an invalidation and the stop order that executes it is covered in what invalidation is and why every trade needs one, and the mechanics of placing the order are in how to set a stop loss that isn't just a guess.
Layer 3 — position sizing, which is just arithmetic
Layers 1 and 2 produce layer 3 automatically. Risk budget divided by the per-unit distance to invalidation gives the size. Nothing here is a judgement call:
- Risk budget = account equity × risk percentage.
- Per-unit risk = distance from entry to invalidation, in the instrument's own units, multiplied by the value of one unit.
- Position size = risk budget ÷ per-unit risk.
A $20,000 account risking 1% has $200. If invalidation sits 25 points away on an instrument worth $2 a point, one unit risks $50, so the position is four units. Change the stop distance and the size changes; the $200 does not. Run your own instrument through the position size calculator, and note that whatever leverage results is an output of this arithmetic rather than a setting you choose — the point of what leverage is and how much is too much.
Layer 4 — daily and weekly caps
Per-trade risk alone permits an ugly outcome: eight consecutive 1% losses in one session is fully compliant with your own rules and costs 8% of the account. Worse, the trades after the third are rarely the same quality as the first three.
- Daily loss cap — commonly two to three times single-trade risk. Hit it and the session is over. Not reduced size, not "one more setup". Over.
- Daily trade cap — a maximum count, which protects quality rather than capital. The reasoning is in how many trades a day a beginner should take.
- Weekly cap — roughly twice the daily cap, to stop four mediocre days compounding into a bad week.
These limits work because they are mechanical. A rule that ends the day removes the decision at exactly the moment your judgement is worst.
Layer 5 — the drawdown rule
The slowest and least dramatic way to lose an account is a long, ungoverned bleed. Layer 5 is a pre-committed response to cumulative loss:
| Account down | Pre-committed response |
|---|---|
| 5% | Review the journal. Look for a pattern before changing anything. |
| 10% | Halve position size until the account makes a new equity high. |
| 15% | Stop trading live. Return at minimum size only after a written review. |
Halving size at 10% is deliberately counter-intuitive. The instinct after a drawdown is to trade bigger to recover faster, which is precisely how a recoverable 10% becomes an unrecoverable 40%. Reducing size costs you very little if the drawdown was noise, and saves the account if it was not.
Deciding this in advance also means you are not diagnosing yourself mid-drawdown, which never goes well. Your trading journal is what turns the review at each threshold into evidence rather than a mood.
What risk management is not
- It is not a prediction tool. None of these layers make you more right. They change what being wrong costs.
- It is not a high win rate. A 40% win rate at 1:3 beats a 70% win rate at 1:0.4 — see the risk-to-reward ratio.
- It is not "trade small and hope". Position size that is too small for your stop distance is its own error; the size should be exactly what the risk budget and the invalidation produce.
- It is not optional above a certain skill level. The traders who last run tighter risk than beginners, not looser.
Frequently Asked Questions
What is risk management in trading?
It is a set of limits decided before you trade that together cap how much any single trade, any single day and any single losing stretch can cost you. The five layers are risk per trade, a written invalidation, a position size derived from those two, daily and weekly loss caps, and a drawdown rule that reduces exposure. A stop loss is only one part of one layer.
How much should you risk per trade?
A fixed small percentage of current account equity, commonly between 0.5% and 2%, chosen so that a realistic losing streak is survivable rather than fatal. Expressing risk as a percentage rather than a fixed cash amount means position size shrinks automatically during a drawdown and grows as the account recovers.
Is a stop loss the same as risk management?
No. A stop loss caps the cost of one trade. Risk management also governs how large that trade is, how many trades you take, what a bad day is allowed to cost, and what happens after a run of losses. A trader with perfect stops and no daily limit can still lose a quarter of an account in an afternoon.
Why do so many retail traders lose money?
Because losses are usually ungoverned rather than unlucky. Regulators have measured the outcome directly: in its 2015 review of retail contract-for-difference providers, the UK Financial Conduct Authority reported that according to its sample data, 82% of clients lost money on these products. Sizing and limits, not forecasting, are what separate the distributions.
Bottom line
Risk management is not the boring part of trading that you get to after the interesting part. It is the part that determines whether you are still trading in a year, which is the only condition under which any edge you develop can pay out. Build the layers in order: a fixed percentage per trade, a written invalidation, a size that falls out of both, a daily cap that ends sessions, and a drawdown rule that cuts exposure before you are desperate. The FCA found 82% of retail CFD clients losing money and responded by constraining exposure rather than teaching analysis — that is the regulator arriving at the same conclusion from the other direction. Start with the 1% rule, and if you are still assembling the wider process, the trading plan guide is where these limits get written down.