When to Cut Losses and When to Let Profits Run

By Lenka Rož Schánová · Operations & Risk, For Traders

Cutting a loss means closing a losing position before it damages your account beyond a pre-defined limit. It is a decision — taken manually or automatically — to accept a small, planned loss now rather than an unplanned, unlimited one later. A stop-loss order is simply the tool that executes that decision for you.

Key takeaways

• A cut loss is the decision to exit a losing trade; a stop-loss order is the resting instruction that carries out that decision without you present.
• Losses compound against you asymmetrically — a 50% drawdown needs a 100% gain just to get back to flat, which is why the size of the cut matters more than the frequency.
• Stop placement should come from structure and volatility (1.5× ATR on XAUUSD, tick-based on US100), not from a round percentage borrowed from an equities article.
• The disposition effect — documented by Terrance Odean in 1998 — explains why traders sell winners and hold losers; pre-set exit rules are the only reliable countermeasure.
• Letting profits run is a mechanical process: a 2R first target, a trailing or Chandelier Exit for the rest, and a time-based exit for trades that stall.
• Inside a prop evaluation, exits are governed by a hard sequence — daily loss limit first, max drawdown second, profit target last.

Watch: related video

What "cut loss" actually means in trading

Cutting a loss is the decision to exit a losing position at a pre-defined level of pain. A stop-loss order is the resting order that carries that decision out without you touching the mouse. That's the whole cut loss meaning in trading — one is a choice, the other is a tool. Most traders blur the two and it costs them.

Cut loss vs stop loss: the difference nobody explains

You can cut a loss without ever having a stop-loss order on the chart — you just close the trade manually because your thesis broke. And a stop-loss order can fire without you making any decision at all — price gaps through it on an NFP print and you're out at a level you didn't choose. Cut loss vs stop loss isn't a semantic argument, it's the difference between judgment and automation, and confusing them is why traders think "I have a stop, I'm protected" when what they actually have is an order sitting at the mercy of liquidity.

Cut loss, take-profit and trailing stop — where each one sits

Think of these three as one exit framework, not competing tools. The stop-loss order caps the downside you're willing to accept. The take-profit order locks in the upside target you calculated before entry — your reward side of the R:R. The trailing stop sits between them: it moves your protective exit in your favor as the trade works, converting an open profit into a locked-in one without you babysitting the screen. Cutting a loss is the mechanism behind the first of these — it's not a separate strategy, it's the discipline that makes the stop-loss order meaningful in the first place.

Why the wording matters when you're building rules

If your trading plan says "cut losses at 1%" but your execution is "move stop to breakeven when price wobbles," you don't have a rule — you have a suggestion. Manual exit vs resting order needs to be an explicit choice in your plan, not something you figure out mid-trade. Across evaluation accounts, the pattern repeats: the trader had a stop-loss order placed correctly, then widened it, deleted it, or hedged around it once price moved against them. The order existed. The decision didn't. That gap — between having a tool and having the discipline to leave it alone — is where most account damage actually happens, long before max drawdown or a daily loss limit ever gets touched.

The maths that makes cutting losses non-negotiable

A 50% drawdown doesn't need a 50% gain to recover — it needs 100%. That single fact, verifiable on any calculator, is the entire argument for cutting losses. The relationship between loss and recovery isn't linear, it's convex, and once you're past a 30% hole the curve goes vertical.

Breakeven recovery: what a drawdown really costs

The breakeven recovery math is simple division: gain required = loss ÷ (1 − loss). Run it across a range and the asymmetry stops being abstract.

Below 20%, the drawdown recovery table looks forgiving — a losing streak feels manageable because the math still roughly matches intuition. Past 30%, it decouples completely. A 50% loss doesn't need a strong quarter, it needs you to double your remaining capital just to get back to zero, before you've made a cent of new gain. This is why every serious risk management framework — including the daily loss limits built into a Trading Challenge — exists to keep you on the flat part of that curve, not the vertical part.

R-multiples and expectancy in plain numbers

R is your unit of risk — the distance from entry to stop, expressed in account currency. Risk $200 to enter a trade, and that trade is worth 1R. A winner that closes at $600 profit on that same setup is a 3R winner. Everything in position sizing and profit cutting gets easier once you stop thinking in dollars and start thinking in R.

Expectancy is where R-multiple and expectancy analysis pays off: (win rate × average win in R) − (loss rate × average loss in R). Take a system with a 35% win rate, a 3R average winner, and a 1R average loser:

(0.35 × 3) − (0.65 × 1) = 1.05 − 0.65 = +0.40R per trade, on average.

That's a positive-expectancy system — one that mints money over a large sample — despite losing on roughly two out of every three trades. Most traders who quit a strategy after eight straight losses were quitting on a system that was working exactly as designed.

Why one 8R winner pays for eight 1R losers

This is the arithmetic behind "let profits run": if your average loser is capped at 1R and one trade runs to 8R, that single winner erases eight full-sized losses and still nets zero — anything beyond that is pure edge. Trend-following and breakout traders lean on exactly this distribution: a low hit rate, small controlled losses, and a handful of outsized winners carrying the whole equity curve.

The condition is non-negotiable: the losses have to stay at 1R. The moment you widen a stop "just this once," an 8R win no longer pays for eight losers — it pays for three, because the other five were 2R or 3R losses in disguise. Letting winners run only works as a system if losses are mechanically, boringly capped every single time.

When to cut losses: a nine-point decision checklist

Cut the trade when any one of these nine conditions fires — you don't need all nine, and waiting for more than one is how a manageable loss becomes an account-damaging one. This is the checklist we'd want pinned above every desk running a cut loss strategy, across any instrument.

Signals that apply to any instrument

1. Thesis invalidated. The reason you entered no longer exists — the breakout failed, the earnings beat didn't hold, the trend line broke. Not "it might come back."
2. Key level closed through. Not wicked through, closed through. Intraday spikes lie; closes tell the truth.
3. Stop distance exceeded by volatility expansion. If ATR has doubled since entry, your original stop no longer represents the same risk — it's now a different, bigger bet you never agreed to.
4. Position size wrong for current ATR. Volatility moved, your size didn't. Cut or trim rather than let the market resize you.
5. Correlated exposure stacking. Long gold, long silver, short DXY all at once isn't three trades — it's one trade three times the size. Cut the weakest leg.
6. News event you didn't plan for. FOMC, NFP, an unscheduled central bank headline — if you didn't size for the event, you don't hold through it.
7. Trade held past its time stop. A swing idea that hasn't worked in the window you gave it is information, not patience.
8. Daily loss limit approaching. This one's mechanical, not emotional — if you're evaluated against a daily loss limit, the checklist item is the account rule itself, not your opinion of the trade.
9. You can't explain why you're still in. The honest one. If you can't state the invalidation level out loud right now, you're not managing risk — you're hoping.

When to cut losses on a stock vs on gold vs on an index future

The checklist is universal; the mechanics of how a stop gets violated are not. When to cut losses on a stock has to account for the gap risk — a stock can close at $50 and open at $44 on a guidance cut, leaping straight past your resting stop with no fill in between. When to cut losses trading gold means respecting how XAUUSD behaves at the London open and around round numbers — $2,650, $2,700 — where liquidity thins right before it floods back in, producing spikes that look like stop hunts and often are. On US100 index futures, the overnight session runs on a noticeably thinner book than regular hours, so a move that would be orderly at 10am New York can slip and gap between 2am and 6am with nobody there to fill you at your price.

The 5–8% rule: useful heuristic or lazy shortcut?

William O'Neil's 7–8% rule — sell any stock that falls 7–8% below your buy point, no exceptions — comes from CAN SLIM, built for unleveraged, cash-account US growth equities where a single security's daily range rarely exceeds 2–3%. Applied there, 7–8% is several multiples of normal noise, a sensible hard floor. Transplant that same fixed percentage onto a leveraged XAUUSD or index futures position and it stops making sense — leverage means an 8% adverse move on the underlying can represent several multiples of your account equity, and you'll get stopped out by normal volatility long before the thesis is actually invalidated. Use O'Neil's rule where it was built to work. Everywhere else, size the stop to the instrument's volatility, not to a number that felt safe for someone else's asset class.

Where the stop actually goes: structure first, ATR second, percentage last

The stop goes at the price that proves the trade wrong — not at a round number, not at a fixed percentage, and not wherever your position size "needs" it to be to feel comfortable. The correct sequence is: find the structural level that invalidates the setup, confirm that level sits at least 1× to 1.5× Average True Range away from entry, then size the position so that distance costs you exactly your per-trade risk budget — never the other way around.

Most traders do this backwards. They decide they're willing to lose $200, then reverse-engineer a stop distance that produces that number, regardless of where the market's actual noise floor sits. That's how you end up stopped out by a normal pullback instead of a real reversal.

A 1.5× ATR stop on XAUUSD around the London open

Say you're long gold off a pullback into the London session. The 14-period ATR on the H1 chart reads 2.80 — that's gold's average hourly range right now, not a guess. The pullback low prints at 2373.40. A round number sits at 2374.00, just above that low, and that's exactly where the crowd parks its stops. Price wicks to 2373.40, sweeps every stop sitting at the round number, then reverses and runs — those traders are out before the real move even starts.

Your stop-loss placement needs to sit below the level the market actually respects, not the level that looks tidy on the chart. Take the pullback low, subtract 1.5× ATR: 2373.40 − (1.5 × 2.80) = 2369.20. That's your XAUUSD stop distance — $4.20 below the low, clearing the sweep zone entirely. It costs more in dollar terms than the round-number stop, but it's the distance that's actually correlated with the trade being wrong, which is the only kind of stop worth paying for.

Tick-based stops on US100 / NQ futures

Index futures don't move in dollars — they move in ticks, and your stop-loss placement has to be translated into tick value before it means anything to your account. NQ (E-mini Nasdaq-100) has a minimum tick of 0.25 index points, worth $5 per tick per contract. A 40-point stop on US100 is 160 ticks. That's 160 × $5 = $800 of risk per contract — before you've even asked whether the setup justifies that stop distance.

This is why "I'll just use a 40-point stop" is meaningless without knowing your contract's tick value. On the Micro E-mini (MNQ), that same 40 points costs $2 per tick — $320 total, a quarter of the full contract's risk. Same structural stop, wildly different account impact, purely because of contract size. Check your specs before you size, every time.

Account size → risk per trade → position size (worked table)

Once you know the stop distance, position sizing is arithmetic, not opinion. Risk budget ÷ dollar risk per unit = your position size. Here's the same 1% risk rule applied across three instrument types:

Notice the gold row leaves $80 of the budget unused and the NQ row leaves $200 — that's intentional. Round down, never up. Sizing to force a "clean" number is exactly the habit that turns a well-placed structural stop into an oversized loss the day the market gaps through it.

Cut losses short, let profits run — the half most traders skip

Letting profits run means mechanically handing your winner to a rule, not to hope. You bank a chunk of the position at a defined multiple of risk to remove the emotional pressure to exit, then trail the remainder with a stop that only moves in your favor — never back toward entry. Cut losses short, let profits run isn't a mantra, it's two separate mechanical decisions, and most traders only ever build the first one.

First target at 2R, then hand the rest to a trailing stop

Close 30-50% of the position once price reaches 2x your initial risk (2R). If you risked $200 on a EURUSD swing, that first partial comes off at $400 in profit. This does three things: it locks in a realized reward, it moves your stop-loss order to breakeven on the remainder risk-free, and it kills the urge to micromanage every tick. What's left runs under a trailing stop with zero cap — this is the piece that pays for the four losers before it.

Trailing stop vs Chandelier Exit vs scaling out

A percentage or fixed-pip trailing stop moves a set distance behind price — simple, but it doesn't adapt to volatility, so it gets whipsawed in quiet markets and too loose in wild ones. The Chandelier Exit fixes that: it trails from the highest high since entry minus 3x ATR (or the lowest low plus 3x ATR for shorts), so the stop widens automatically when volatility expands and tightens when it contracts. Scaling out — taking 25% off at 2R, another 25% at 4R, trailing the rest — works best when you genuinely don't know if you're in a trend leg or a fakeout; it hedges your uncertainty about the regime rather than betting the house on one exit method.

The time-based exit: when a trade stops earning its margin

A setup can be technically "still valid" and still be dead. If your swing-trade thesis expects a move within 3-5 trading days and you're flat after day 5 with the stop untouched, that's your time-based exit — close it, don't wait for the stop to prove you wrong. Capital sitting idle in a stalled trade is capital not working an ATR-based setup elsewhere; margin has an opportunity cost even when the position isn't technically losing.

Trending vs choppy: a regime-based exit matrix

Match the tool to the tape, not the tool you're most comfortable using — a Chandelier Exit in a choppy range just donates your open profit back to the market one pullback at a time.

The stop-loss and take-profit mistakes that quietly drain accounts

Most account damage doesn't come from being wrong about direction — it comes from where the stop and target were placed relative to how the instrument actually moves. The same five errors show up across forex, gold, indices, and futures accounts, and each one has a mechanical reason it fails, not just a label.

Round-number stops and the liquidity sitting on top of them

Stops parked at 1.1000, 2000.00 on gold, or a clean 20,000 on an index aren't random — they're where thousands of other retail stops sit too, and that clustering is exactly what makes them attractive to fill against. Round-number stop hunting isn't a conspiracy theory, it's liquidity mechanics: price needs orders to fill into, and a shelf of stops one tick below a round number is an easy pool to sweep before reversing. Put your stop a few pips or a few ticks off the round number, behind the actual structure — the last swing low, the ATR-based buffer — not on it.

Moving the stop, mental stops and other forms of negotiation

We've all slid a stop hoping price comes back. The MAE (maximum adverse excursion) data on those trades usually says it doesn't — you're not managing risk at that point, you're negotiating with a chart. Moving your stop loss away from price converts a planned 1R loss into an unplanned 3R accident, and it's the single fastest way to turn a manageable drawdown into an account-ending one. A mental stop loss — "I'll close it if it hits X" without an actual order in the book — fails for a related reason: it depends on you having full attention and a clear head exactly when volatility spikes, which is precisely when you don't. NFP prints and FOMC releases don't wait for you to open the platform. Put the order in the system.

Targets set inside the instrument's noise band

A take-profit placed inside the average bar range gets chopped out before the move even develops. If XAUUSD is averaging 12-15 points of noise per hour and your target sits 6 points from entry, you're not trading a thesis — you're gambling on which side of the chop you land. Measure the instrument's ATR first, then set targets that live outside that noise band, giving the move room to actually express itself.

There's a sizing mistake hiding underneath all of this that nobody frames as an exit problem: an oversized position forces a stop that's tighter than one ATR just to keep the dollar risk survivable. That's backwards. The instrument's volatility should set the stop distance; your position size should shrink to fit that stop, not the other way around. Get the sizing right first, and half of these "stop loss and take profit mistakes" stop happening on their own — profit cutting becomes a deliberate choice again, not damage control.

Pre-set exit rules: how to take the decision out of the moment

The fix for bad exits isn't more willpower in the moment — it's writing the exit rules before you enter, so there's no decision left to make when the trade is live. Every rule you write in cold blood before a position exists beats every rule you'd try to invent while that position is bleeding.

The disposition effect: why you sell winners and hold losers

Terrance Odean's 1998 study of over 10,000 retail brokerage accounts documented what's now called the disposition effect: traders realise winning positions at a materially higher rate than losing ones, closing gains early and dragging losers far past where a rational rule would have cut them. It's not stupidity — it's wiring. Kahneman and Tversky's prospect theory explains the asymmetry underneath it: a loss hurts roughly twice as much as an equivalent gain feels good, which is exactly why you grab a small profit to feel the relief and hold a loser hoping it turns around before you have to feel the pain. Loss aversion isn't a character flaw you can think your way out of mid-trade — it's a default setting that shows up in the data every single time.

Writing exit rules before entry, not during the trade

The countermeasure is mechanical, not motivational. Before you click buy or sell, your trade plan should already specify:

• Stop level — the price that invalidates the setup, set by structure or ATR, not by how much dollar risk feels comfortable.
• First target — where you take partials or bank the trade, defined in R multiples, not "when it feels right."
• Trail rule — the specific mechanic (structure break, moving average, ATR trail) that lets a winner run without you improvising an exit on gut feel.
• Time stop — the point where you exit flat if the move hasn't developed, because a stalled trade is capital doing nothing.
• The early-cut condition — one line naming exactly what news, level break, or invalidation would make you exit before your stop is even touched.

Write these into your trading plan and you've already made the decision — the trade just executes it later.

What Steenbarger and O'Neil actually said about exits

William O'Neil, founder of Investor's Business Daily, built an entire methodology around one blunt rule: cut every loss at 7-8% below purchase, no exceptions, no stories. His logic was arithmetic, not sentiment — a 50% loss needs a 100% gain just to break even, so you never let a small loss become that math problem. Brett Steenbarger, the trading psychologist who's coached hundreds of professional traders, frames rule-following itself as a trainable skill, not a fixed personality trait — something you build the same way you build any other competence, through repetition under real stakes. That's worth sitting with: you're not born disciplined, you practice it, the same way you'd study trading psychology to understand why the rule feels hard to follow in the first place. Discipline isn't the absence of the urge to move your stop — it's following the pre-written rule anyway.

Exit rules inside a prop firm evaluation

Inside a prop firm evaluation, three limits govern every decision you make, and they don't carry equal weight. The daily loss limit stops your session, the maximum drawdown stops your account, and the profit target is the only one of the three you're allowed to take your time with.

Sequencing daily loss limit, max drawdown and profit target

Treat them as a hierarchy, not a checklist. Breach the daily loss limit and you're done for the day — no more entries, no "one more trade to get it back." Breach max drawdown and the evaluation account is closed, full stop. The profit target has no clock attached beyond the challenge's overall time window, which means it's the variable you can flex when the other two are under pressure. A trader who protects the daily loss limit and the max drawdown with religious consistency will eventually reach the profit target almost as a byproduct. A trader who chases the profit target first and treats the loss limits as an afterthought is the one who blows the account on day four.

Setting a personal daily cap tighter than the platform limit

Reverse-engineer your per-trade risk from the daily loss limit before you take a single entry. If the platform's daily cap allows for three losing trades at your normal stop size, that's your real risk budget for the day — not a suggestion, a ceiling. Then cut it further: run your own personal daily cap at 60-70% of the platform's number. A bad tick, a slippage event on a news spike, or a fat-fingered lot size then hits your personal wall long before it touches the account-ending one. This is the same logic behind the For Traders challenge rules — the rules are the floor, not the target.

Breaking the profit target into weekly goals

Splitting the profit target into weekly milestones keeps you from getting cornered into oversized positions in the final days of the evaluation. A trader with 80% of the target banked by week three trades normally. A trader who's flat by week three and has one week left starts widening stops and doubling size — exactly the behavior that turns a manageable drawdown into a blown account. Set a weekly number, and if you miss it, extend your timeline rather than your risk.

All of this runs on simulated capital during the evaluation, which is precisely why the evaluation is the right place to test these rules — you get to find out how you behave under a daily loss limit before real performance rewards are on the line. That's true whether you're running the Two-Step Challenge, which gives you two phases to prove the same discipline twice, or Instant Funding, where there's no evaluation phase and the max drawdown rule applies from day one on a funded account.

Audit your exits: what MAE and MFE tell you

Maximum adverse excursion (MAE) is the worst unrealised drawdown a trade hit before you closed it; maximum favourable excursion (MFE) is the best unrealised gain it reached before you closed it. Track both on every trade and your exits stop being a feeling — they become a dataset you can actually diagnose.

Most traders can tell you their win rate and their average R:R. Almost none can tell you their average MAE on winners versus losers. That gap is exactly where exit quality hides. MAE MFE analysis takes the guesswork out of "was that stop too tight" or "did I cut a winner early" and turns it into a number you log next to every trade.

Reading maximum adverse excursion: are your stops too tight?

Pull your last 20 losing trades and check where price was relative to your stop when it eventually would have hit target — if it ever would have. If a cluster of your losers got stopped out within a tick or two of their MAE, then reversed and ran to where your target sat, that's not bad luck, that's a stop placed inside the instrument's normal noise. On XAUUSD during London/NY overlap, a stop tighter than 1.2–1.5x ATR(14) gets clipped by ordinary chop, not by the market proving you wrong. If this pattern shows up in more than a handful of your 20 sampled trades, the fix isn't a new entry system — it's a wider stop and a smaller position size to keep the dollar risk identical.

Reading maximum favourable excursion: are you cutting winners early?

Now flip it and look at your closed winners. If MFE on those trades regularly ran to 2R or 3R but you closed at 1R, you're leaving the outlier on the table trade after trade. This is cutting winners early, and it's the quieter cousin of cutting losses too late — both come from the same discomfort with an open position moving around. One or two early exits don't matter. A pattern across 15 of 20 trades means your take-profit logic, or your nerve, is capping upside that the market was willing to give you.

The 20-trade review loop

Log entry, stop, target, MAE, MFE, exit reason, and market regime (trending, ranging, news-driven) for every trade. Review in batches of 20 — enough to see a pattern, not so many that you're reviewing decisions you've already outgrown. Change exactly one variable per review cycle: stop distance, target distance, or exit trigger — never all three at once, or you won't know which change moved the needle. This is the same discipline a proper trading journal review enforces automatically if you're already logging trades there.

The honest trade-off: a strict cut-loss rule

Pros

• Caps the worst single trade at a known number, which keeps the breakeven maths survivable
• Removes in-trade negotiation — the decision is already made before the candle prints
• Makes R-multiples comparable across instruments, so expectancy becomes measurable
• Keeps you inside a daily loss limit and max drawdown without last-minute improvisation
• Frees mental bandwidth for entry quality instead of damage control

Cons / risks

• Rigid stops get taken out by normal volatility if the distance isn't matched to the instrument's ATR
• A high stop-out rate raises the cost of spread, commission and slippage over a sample
• Cutting on a mechanical rule occasionally exits a valid thesis one bar before it works
• Requires you to trust the sample over the individual trade, which is psychologically harder than it sounds
• Poorly calibrated rules can encourage over-trading as you chase back the small losses

Frequently Asked Questions