Guide · Updated 5 August 2026

Risk Management and Position Sizing: The Full Playbook

Almost nobody blows up an account because their entries were bad. They blow up because the size was wrong. This is the arithmetic that decides whether a normal losing streak is an inconvenience or the end of the account.

Why sizing beats picking

Two traders take exactly the same ten signals. Both are right six times and wrong four times. The first risks a fixed 1% of equity per trade and takes profit at twice the risk. The second sizes by gut feel — small when nervous, huge when convinced — and happens to be convinced on three of the four losers. The first trader finishes the sequence up roughly 8%. The second finishes down. Same edge, same signals, opposite outcome.

That is the whole argument for treating risk as a system rather than a mood. An edge only pays out over a large sample, and you only get a large sample if you are still solvent. Every rule below exists to keep you in the sample.

Rule 1: fix the dollar risk before you look at the price

Fixed fractional risk means deciding, in advance, that any single trade may cost you no more than a set percentage of equity. On a $10,000 account at 1%, that is $100. Not $100 of capital deployed — $100 of loss if the stop is hit.

The order of operations matters. Amateurs pick the position size first ("I'll buy 100 shares") and discover their risk afterwards. Professionals fix the risk first and let the chart decide the size:

Position size = (Account equity x Risk %) / (Entry price - Stop price)

Worked example. Equity $10,000. Risk 1% = $100. You want to buy a stock at $48.20 and the structure says the idea is wrong below $45.80. Stop distance is $2.40. Position size is 100 / 2.40 = 41 shares, about $1,976 of capital. If instead the stop sits at $47.20 — a tighter $1.00 away — the same $100 of risk buys 100 shares, roughly $4,820 of capital. The risk did not change. Only the exposure did.

This is why "how much should I buy" is an unanswerable question until the stop is chosen. The free position size calculator does this arithmetic for you, including for fractional and crypto units.

Rule 2: put the stop where the idea dies, not where the pain starts

A stop-loss is not a pain threshold. It is the price at which your reason for being in the trade no longer holds. Three defensible placements:

• Structural. Below the swing low that defines the uptrend, or above the swing high for a short. If that level breaks, the trend read was wrong.
• Volatility-based. 1.5x to 2x the 14-period Average True Range away from entry. This adapts automatically: a $400 index name and a $6 small cap get very different distances.
• Level-based. Under VWAP for an intraday reclaim trade, under the moving average that the trade is anchored to, or under the breakout level that triggered entry.

The common failure is the arbitrary percentage stop. A 5% stop on a stock with a 6% average daily range is not risk control — it is a coin flip that pays the spread. Check ATR before you choose, and if the honest stop is so far away that the position becomes tiny, that is information: the trade is being taken at a bad location.

Rule 3: demand a reward-to-risk floor

Reward-to-risk (R:R) is the distance to your target divided by the distance to your stop. It determines what win rate you need just to break even:

• 1:1 — you need to win more than 50% of the time.
• 1.5:1 — you need 40%.
• 2:1 — you need 33.3%.
• 3:1 — you need 25%.

Those figures are before commissions, spread and slippage, which in practice add a few percentage points to the required win rate. This is why SIGNAL9's conviction gate refuses to publish swing setups below roughly 1.8:1 measured against the buy zone: a 1:1 trade needs near-perfect selection to survive costs, and no scanner is that good.

Be honest about the target. The correct target is the next real supply level — prior resistance, a measured move, the top of the channel — not whatever number makes the ratio look acceptable. Inflating the target to justify a trade is the most common way traders lie to themselves with a spreadsheet.

Rule 4: budget total open risk, not just per-trade risk

One percent per trade sounds conservative until you hold eight positions and the whole market gaps down together. Add a portfolio-level cap: total open risk of 4–6% of equity, and no more than about 2% concentrated in a single theme.

Correlation is the trap. NVDA, AMD, TSM and a semiconductor ETF are not four trades; on a risk-off morning they are one trade held four times. The same goes for a book of high-beta crypto alts, which tend to move as leveraged expressions of Bitcoin. Group your positions by what actually drives them — AI infrastructure, rates, oil, BTC beta — and cap the group.

Rule 5: understand drawdown maths before you need it

Losses and recoveries are not symmetric. Down 10% needs +11.1% to get back. Down 25% needs +33%. Down 50% needs +100%. Down 75% needs +300%. Every extra percent of risk you take today buys a disproportionately harder recovery later.

At 1% risk per trade, ten consecutive losses cost roughly 9.6% of equity — unpleasant, fully recoverable, and statistically normal for a 45%-win-rate system. At 5% risk per trade, the same losing streak costs about 40%, which requires a 67% gain to undo. Nothing about the strategy changed. Only the size did.

Rule 6: manage the trade with rules, not with feelings

Decide the exit plan before entry and write it down. Three patterns worth knowing:

• Scale at 1R. Sell a third of the position once price has moved one unit of risk in your favour, and move the stop to break-even on the rest. This converts a live risk into a free option and dramatically smooths the equity curve.
• Trail behind structure. Move the stop under each new higher low rather than a fixed percentage. It keeps you in trends without giving back the whole move.
• Time stop. If the thesis was "this breaks out within five sessions" and it has done nothing in five sessions, the thesis expired. Close it and free the risk budget.

What you should not do is widen a stop because price is approaching it. Moving a stop away from entry converts a planned small loss into an unplanned large one, and it is the single most expensive habit in retail trading.

Rule 7: keep a record that can prove you wrong

Log every trade with entry, stop, target, size, the reason, and the outcome in R multiples rather than dollars. Thinking in R — "that was a +2.3R trade" — strips out account size and makes performance comparable over time. After fifty trades you can answer real questions: which setup actually pays, whether your losers are bigger than planned, whether your best results come from a single ticker you got lucky in.

This is the same discipline we hold ourselves to publicly. Every ATLAS trade is written to the bot track record at entry with its target and stop already visible, and losing trades stay on the record permanently. If a track record only shows winners, it is marketing, not evidence.

A five-line pre-trade checklist

1. Where does this idea become wrong? That is the stop.
2. Where is the next real supply level? That is the target.
3. Is the ratio at least 1.8:1 after costs? If not, skip it.
4. What size makes the stop cost exactly 1% of equity?
5. Does this push total open risk or theme risk over the cap?

Five questions, under a minute, and they remove almost every catastrophic outcome available to a retail trader. The edge is optional. The arithmetic is not.

FAQ

Disclaimer: educational content only, not investment advice. See our disclaimer.