How to Build a Prop Firm Risk Management Plan Before Your First Trade
Key takeaways
- Your risk per trade is an output, not an input. It falls out of two decisions: where your own daily stop sits below the firm's, and how many consecutive losses the plan has to absorb before it reaches that stop.
- Put the personal stop below the firm's limit so a breach needs two failures instead of one. A plan that ends exactly on the firm's number leaves nothing for slippage, a widened spread or a gapped fill.
- The drawdown mechanic changes the whole calculation. Under a static floor, realised profit widens your buffer. Under a trailing floor the buffer follows your peak, so an account can be green and have less room than it started with.
- Positions that lose together are one position. Three trades on the same directional theme, each carrying the plan's full unit, quietly triple that unit.
- The daily loss limit resets overnight. The maximum drawdown does not, so the plan needs a written rule that cuts exposure as the total buffer shrinks.
- Every figure below is a worked example on hypothetical numbers. Prop firm rules are commonly structured in similar shapes, but they differ by firm, account type and platform, and they change, so take the real figures off your firm's own site.
A prop firm risk management plan is arithmetic, not preference
A prop firm risk management plan is not a preference. It is arithmetic derived from two numbers the firm already published: the daily loss limit and the maximum drawdown on the account you bought. If you failed an evaluation and suspect sizing rather than signal, this is usually where it happened: the risk per trade got chosen first, the limits got checked afterwards.
The order matters because the two constraints are not symmetrical. A profit target is a number you can miss and keep trading. A loss limit closes the account the moment it is crossed. So the plan starts at the loss limits and works backwards to a lot size.
Everything below is a worked example on hypothetical figures. Prop firm rules are commonly structured in similar shapes, but they differ by firm, account type and platform, and they change. Take the real numbers off your firm's own site, note the date, and rerun this with yours.
Deriving your risk per trade from the daily loss limit
Most traders start from a percentage they absorbed somewhere: one percent, two percent, half a percent. That number is the output, not the input. Two decisions produce it: where your own daily stop sits below the firm's, and how many losses in a row the plan must absorb before reaching it.
Worked example, hypothetical numbers. Take a 100,000 account. Assume a daily loss limit commonly structured at around 5 percent, so 5,000, and a maximum drawdown commonly structured at around 10 percent, so 10,000. Verify both against your own account first.
Step one: put your personal daily stop at 60 percent of the firm's limit: 3,000, or 3 percent of the account. The gap is not timidity: slippage, a widened spread and a gapped stop are all real, and a plan ending exactly on the firm's number has no margin for a bad fill.
Step two: decide the streak. Say the plan must absorb six full losses in a row before reaching that stop. Then the risk per trade is 3,000 divided by 6, which is 500, or 0.5 percent of the account.
Now check that against the firm's own limit. At 500 a trade it takes ten full losses to reach 5,000, and your stop arrives at six. Those four losses of daylight, worth 2,000, are what make a breach need two failures, not one. Our note on [position sizing for a challenge](/blog/position-sizing-half-percent-rule) runs the same logic.
- Firm daily loss limit: 5,000 on a 100,000 account.
- Personal daily stop, 60 percent of it: 3,000.
- Consecutive full losses the plan must absorb: 6.
- Risk per trade: 3,000 divided by 6, so 500, or 0.5 percent.
- Personal stop to firm limit: 2,000, another four full losses.
How many losses in a row the plan has to survive
The streak is the input almost nobody writes down, and it decides whether a normal bad run is survivable or terminal. Any method that loses a meaningful share of its trades can produce runs of four, five and six losses, and they do not arrive politely spaced out.
Choose the streak honestly: pull your last hundred trades, find the longest run of losses, then add two. You are sizing for the worst week you have already had.
Getting it wrong costs you immediately. Same worked example: at 2 percent, one loss of 2,000 leaves only 1,000 under the 3,000 personal stop, so the plan cannot take a second full position at all. That is not bad luck and not a bad entry. The strategy did not fail, the plan did.
Here is the same hypothetical day at three risk settings, on that 100,000 account with its 5,000 limit and 3,000 personal stop.
- 0.5 percent, 500 a trade: six full losses reach the 3,000 stop, ten reach the firm's 5,000.
- 1 percent, 1,000 a trade: three losses reach the personal stop, five reach the firm's limit.
- 2 percent, 2,000 a trade: one loss leaves 1,000 under the personal stop, half a position. Override that stop and two losses total 4,000, a third 6,000, past the firm's limit.
Why static and trailing drawdown produce different plans
The daily limit decides the size of a bad day. The maximum drawdown decides how many bad days the account survives, and its mechanic decides whether profit buys room.
Worked example, static floor, hypothetical numbers. A 100,000 account with a 10 percent static maximum loss has a floor fixed at 90,000 for the life of the account. Total room is 10,000: twenty full losses at 500 a trade, or three complete stop days of 3,000 with 1,000 left over. Realised profit widens that gap permanently, so at 104,000 the distance to the floor is 14,000.
Worked example, trailing floor, same hypothetical account. The floor follows new highs and never comes back down. The account peaks at 105,000, which lifts the floor to 95,000, and then the balance gives back to 101,000. That is 1,000 of profit, and a static floor at 90,000 would leave 11,000 of room, but the trailing floor is stuck at 95,000, so the real buffer is 6,000. Green, with less room than you started with.
So under a trailing floor the risk per trade is not set once at the start. It is recalculated from the live distance to the floor. At 500 a trade a 6,000 buffer is twelve full losses of room; a trader who reads the profit line and doubles to 1,000 has six. The mechanics are in [trailing drawdown explained](/blog/trailing-drawdown-explained), and the interaction of both limits in [daily loss versus max drawdown](/blog/daily-loss-vs-max-drawdown). Confirm four things in the firm's current documentation.
- Is the maximum drawdown static under the starting balance, or does it trail upward?
- Is it measured on closed balance or live equity, and does floating profit move it?
- If it trails, does it stop trailing at any point, such as the starting balance?
- When does the daily limit reset, in which timezone, and from balance or equity?
Correlated positions count as one risk
The plan says 500 a trade. Three positions open at once, each risking 500, is not three separate risks if all three are the same bet wearing different tickets.
Worked example, hypothetical numbers. Long EURUSD, long GBPUSD and short USDCHF are three ways of being short the dollar. One dollar move against you sends all three toward their stops at once, so the honest worst case is 1,500. That is 1.5 percent of the account and half the 3,000 personal stop, in what felt like three small decisions.
Correlation is not only about pairs, it is about the event. Three positions that all lose if one inflation print lands wrong are one position, whatever instruments they sit in.
Write the rule as a number: one unit of risk per directional theme. If the theme is worth 500 and you want three legs, they share the 500, roughly 167 each, not 500 each. Exposure caps are commonly structured as a rule of their own, separate from the loss limits, the subject of our piece on [max lot size limits](/blog/prop-firm-max-lot-size-limits).
- Group open positions by what would have to be true for all of them to lose at once.
- Cap simultaneous positions as a plain number, because a rule you must reason about at speed is a rule you will skip.
What the plan does after a losing day
This is the clause that gets improvised, and improvising it is how one red day becomes a failed evaluation. The rule is short: the daily loss limit resets overnight, the maximum drawdown does not.
Worked example, hypothetical numbers, static floor. One full personal stop day costs 3,000. Tomorrow's daily limit is a fresh 5,000, but the 10,000 total buffer is now 7,000. A second stop day takes it to 4,000, so 60 percent of the account's lifetime buffer is gone after two sessions that both obeyed the daily rule.
One more reason to write this part carefully. Only 1 to 3 percent of funded traders keep the account long term, and the give-back usually does not happen during the challenge. It happens later, on the first genuinely bad week, with a plan that was never written down, a pattern covered in [why funded traders lose the account](/blog/why-funded-traders-lose-the-account).
So the plan needs clauses that shrink exposure as the buffer shrinks, each written as a number.
- Hit the personal stop and the platform closes for the day. No last setup to win it back.
- Trade the next day at half size, 250 here, until one green day restores the normal number.
- Two stop days in a row: nothing more that week, plus a written review.
- Recompute risk from the remaining buffer, not the starting balance, every time the floor moves.
- Never increase size to recover lost time. A missed target costs a fee, a breached limit costs the account.
Write the plan down before the first trade
All of this is worthless as a memory: the worst moment to compute a lot size is the moment you most want to be in a trade. Write the page while nothing is at stake, and obey it while something is.
A risk per trade number assumes every position carries a stop from the instant it opens, because without one the figure is a guess, not a limit. Some firms require a stop on every trade and some do not, but both measure you against the same drawdown. Our note on [mandatory stop loss rules](/blog/mandatory-stop-loss-prop-firms) covers how those clauses are worded.
These are the lines it has to carry.
- Firm, account size, phase, and the date I last read the rules.
- The daily loss limit, how it is measured, when it resets and in which timezone.
- The maximum drawdown, static or trailing, on balance or equity, and its lock point.
- My personal daily stop as a number, the consecutive losses the plan must absorb, and the risk per trade that follows.
- Maximum simultaneous positions, and the correlation groups I treat as one risk.
- What I do after one stop day, after two, and how normal size is earned back.
Enforcing the plan when you least want to
The last problem is enforcement, which is not arithmetic. A plan you can override at two in the morning is a preference again. This is where a rules-aware tool has an honest place: it holds the numbers you derived and acts on them without asking. PraxAI, which publishes this blog, ships PraxAI GUARD for that job: it is built to flatten and lock the account before a loss limit is breached, not at it. No tool passes an evaluation for you, no software removes market risk, and nothing here is a promise of a pass or of income. Our criteria for judging any [AI trading bot for prop firms](/blog/best-ai-trading-bot-prop-firms-2026) are worth reading first.
None of this makes you a better trader. It makes the account's arithmetic survivable long enough for whatever edge you have to appear. Repeat challenge fees can add up to $2,400+ a year, and most of that spend buys the same lesson twice. Write the page first, then take the first trade.
Frequently asked questions
What is a prop firm risk management plan?
It is a written document that converts the firm's published limits into your own trading numbers before you place a trade. At minimum it states the firm's daily loss limit and maximum drawdown, your own daily stop set below the firm's, how many consecutive losses the plan must absorb, the risk per trade that follows from that, how correlated positions are counted, and what happens after a losing day. Every number is derived rather than chosen, so none of them has to be decided while you are losing.
what should my risk per trade be if my prop firm daily loss limit is 5 percent?
Divide, do not guess. Worked example on hypothetical numbers: on a 100,000 account a 5 percent daily limit is 5,000. Set your own stop below it, say 3,000, then decide how many losses in a row the plan must survive, say six. That gives 3,000 divided by 6, which is 500 per trade, or 0.5 percent. Change either input and the answer changes, which is the point. Check the actual limit for your own account type, since these figures are illustrative only.
Does a trailing drawdown change how I should size positions?
Yes, substantially. Under a static floor, realised profit permanently increases the distance to your breach level, so the buffer you size against grows as you win. Under a trailing floor the level follows your peak, so the buffer stays roughly constant while it trails and can be smaller than your starting buffer even when the account is in profit. The practical rule is to size from the live distance to the floor rather than from the starting balance, and to recompute it after every new high.
How do I count correlated trades against my risk limit?
Treat a correlated cluster as one position. Group trades by what would have to be true for all of them to lose together, whether that is the same currency on one side, the same directional theme, or the same scheduled news release. Then allocate one unit of risk to the group and divide it across the legs. Three trades on one theme, each carrying the plan's full unit, is triple risk described in single risk language.
What should I do the day after I hit my daily stop?
Trade smaller, and write the rule before you need it. A common structure is to halve the risk per trade the next session and restore normal size only after a green day, then stop for the week entirely after two stop days in a row. The reason is arithmetic rather than psychology: the daily limit resets overnight but the maximum drawdown does not, so each stop day permanently reduces the buffer the account has left for the rest of the evaluation.
Want the bot that runs this discipline for you?