How Much Should You Risk Per Trade? A Decision Guide

“Risk 1% per trade” is the most repeated number in trading, and for most people it is a fine default. But a default is not a decision. The right per-trade risk for you depends on how your system actually behaves and how much room your account gives you before it fails.

Here is a framework for deriving that number from three inputs instead of inheriting it from a forum post.

Why one risk number doesn’t fit every trader

Two traders can both “risk 1%” and be running completely different levels of danger. One takes three trades a week on a high win-rate mean-reversion system; the other fires twenty correlated scalps a day into the same instrument. The percentage is identical; the exposure is not.

A per-trade risk number only means something in the context of how often you trade, how correlated your trades are, and how far your account can fall before it breaks. Copying a percentage without those three things is copying the answer to someone else’s exam.

The framework below reduces to three inputs. Get honest numbers for each and the risk percentage almost falls out on its own.

Input 1: your measured win rate and payoff

You cannot size a system you have not measured. The first input is your real, historical win rate and payoff ratio — average winner divided by average loser, ideally expressed in R-multiples so it is independent of position size.

A high win rate with small winners tolerates a different risk profile than a low win rate with large winners.
What matters is the combination, and the combination is captured by your expectancy — the average amount you expect to make per trade, per unit risked.

Run a batch of your closed trades through an expectancy calculator before you touch the risk question. If your expectancy is not clearly positive, the correct risk per trade is zero — no sizing scheme fixes a negative edge, it just changes how fast you lose. Everything downstream assumes you have first confirmed the edge is real.

Input 2: the max drawdown your account can take

The second input is a hard external limit: how far can your equity fall before something bad and irreversible happens? On a personal account that is your own pain threshold. On a prop-firm account it is a rule — a maximum drawdown, often trailing, that ends the account the moment it is breached.

Two practical points:

Prop firms move the goalposts. Whether the drawdown is calculated on balance, equity, or a trailing high-water mark — and the exact percentage — varies by firm and program. Always confirm the current rules directly with your firm rather than trusting a number you read once.
Model it before you trade it. A prop-firm drawdown calculator lets you see how a string of losses at a given risk percentage walks your equity toward the breach line, especially when the limit trails your peak.

Your per-trade risk has to be small enough that a realistic cluster of losses — not just one — stays comfortably inside this budget.

Input 3: your emotional tolerance for a losing run

The third input is the one spreadsheets ignore. Even a genuinely profitable system delivers losing streaks that feel like the edge has vanished. The question is not whether you will hit a rough patch; it is whether you will keep executing the plan when you do.

Be brutally honest here. If a run of six or seven losses in a row will push you into revenge-sizing, moving stops, or skipping valid setups, then your behavioral max risk is lower than your mathematical max risk — and the behavioral one wins, because a plan you abandon has no expectancy at all. Size to the version of you that shows up on a bad week, not the version that backtests calmly on a Sunday.

Putting them together into a defensible risk %

Now combine the three. The logic is a filter, and you take the smallest number that survives all three gates:

Edge gate. Confirm positive expectancy. If not, stop — risk is zero.
Survival gate. Pick a risk percentage where a plausible worst-case losing streak stays inside your drawdown budget with margin to spare. Fewer, more correlated trades demand a smaller number.
Behavior gate. Lower it further if that streak would break your discipline.

For most traders the survivable answer lands somewhere modest — often at or below 1% — precisely because the drawdown and behavior gates bite before the growth-maximizing gate does. Once you have a number, an position size calculator turns it into an exact contract or lot count for each trade’s stop distance, so “risk 0.75%” becomes a concrete order rather than a vibe.

The hard part is not the arithmetic — it is holding the line trade after trade. This is where enforcement beats intention. Shibiki lets you push a per-trade and per-day risk ceiling as a hard limit at the broker, so the account itself refuses to exceed the number you committed to on a calm day. The decision you make with this framework becomes a rule the market cannot argue you out of.

When to dial risk up or down as conditions change

Your risk percentage is not a tattoo. It should breathe with the evidence:

Dial down after a drawdown, when moving to a new instrument or session, or while your sample on a strategy is still thin and your win-rate estimate is noisy.
Dial up — gently — only when a large, stable sample confirms the edge and your equity has room, never as a reaction to a hot streak.

Shibiki’s live edge health helps here by wrapping each strategy’s win rate in a Wilson confidence interval: when the interval is wide, you are guessing, and guessing is a reason to size down. Let the data widen your risk, never your feelings.