Should You Take Profit Early or Let Your Winners Run?

On a breakout system, taking profit early is not a safety net. It is a cost. On my own gold systems I tested more than twenty alternative exits against the existing one, including several ways of banking part of a winner early, and none of them beat letting the whole position run. Banking early made the result worse and left the drawdown deeper, not shallower.

MasterVP · Systematic Trading Journey

Written by the founder of KenKem, a software engineer of twenty years who traded gold by hand before building the research stack behind MasterVP. This is for anyone who has closed a winning trade early and then watched it run without them, which used to be me most weeks. It is knowledge sharing, not a recommendation. Every figure below is a backtest or a research measurement on real tick data, from my release 1.26 write up of 15 September 2026 and the study notes behind it, and none of it is a live result.

Why does banking a winner early feel like the safe choice?

Because a profit you have not closed feels like a profit you can still lose, and a loss feels heavier than the same gain.

When I traded by hand, the moment a trade went green my attention changed. I stopped asking whether the move was finished and started asking how much of it I could keep. Closing early felt responsible. It also had a name in the research literature long before I felt it, the disposition effect: people sell their winners too soon and hold their losers too long, because realising a gain feels good and realising a loss feels like admitting a mistake.

The problem is that the feeling judges one trade at a time. A system is judged over hundreds. A decision that feels prudent on Tuesday afternoon can be quietly expensive across a year.

Where does the result of a breakout system actually come from?

From a small number of large winners that pay for everything else.

On my shipped five-minute gold system, over 2024 to 2026, the best 30 of 1,643 trades add up to about three quarters of the whole result. The best 100 add up to more than twice the whole result, which means the other 1,543 trades, taken together, lose money. One trade that reaches its full target pays for about five full losses.

That is not a flaw I am confessing. It is how a breakout system is supposed to look. Most trades either stop out or get scratched near break-even, and a few run far enough to carry the rest. A breakout system that did not look like that would usually be cutting its winners short. I wrote about why that concentration makes a skipped trade so expensive in should you trade a system by hand or automate it.

So what does taking partial profit actually sell?

It sells a slice of the rare big winner to buy comfort on the ordinary trade.

Here is the trade-off in one line. A partial close shrinks the position on every trade that reaches the first milestone, including the handful that would have run to the far target. The full-size losers are untouched, because they never reached the milestone. So you cut the exact trades that pay for the losses and leave the losses as they were.

In August 2026 I tested it directly, against the shipped exit, which banks nothing early. Banking a fixed share of the position early, at several sizes up to half, lowered the result at every size, and the drawdown came out deeper than simply holding at every size too. That second part surprised me. I expected banking to at least smooth the curve. It did the opposite, because the winners that used to offset a bad run were now half the size when the bad run arrived.

A month later I tested a different shape, pre-registered this time: bank a small slice and use it to justify a wider stop on the rest. Across slices from one percent to half, the bigger the slice the more it hurt, with no sweet spot in between, and on honest walk-forward selection the chosen version still lost to simply holding, in every fold on the five-minute system.

Banking early is a claim on the average trade, paid for by the rare one.

What about cutting a trade early when it starts to look wrong?

I tested that too, and it fails for the same reason.

The idea was to close a breakout trade as soon as price slipped back into the recent local value area, before the stop was hit. It sounds like obvious risk management. Across 32 tested variants all 32 were negative. Every time the rule cut a trade that would have lost, it saved one unit of loss. Every time it cut a trade that would have recovered, it gave up between roughly two and six units. A cheaper version that only tightened the stop looked positive in a quick bar-level estimate, and turned negative when it was replayed on real ticks.

Other early exits went the same way: exits keyed to a moving average turning, to price reverting toward its mean, to momentum fading. None cleared the bar set in advance.

How does my system handle a winner instead?

It decides in advance and does not ask me in the moment.

Once a trade has moved far enough in its favour, the stop moves to break-even, and from there a trailing stop follows price and only ever tightens. The whole position stays on. Either the move exhausts and the trail closes it, or it reaches a far target. I do not publish the distances, but the mechanism is the point. Every exit decision is made by a rule written before the trade, so the version of me that wants to bank because the last three trades lost never gets a vote.

That is the part that removed the emotion for me. Not a better feeling about open profit, which I still have, but the fact that the feeling no longer reaches the order.

What does letting winners run cost you?

Quite a lot, and none of it is hidden in the numbers above.

Long stretches of scratches and small stops while you wait for the one that runs. A real chunk of every big winner handed back on the way out, because a trailing stop only knows the move is over after it is over. Losing streaks that are ordinary arithmetic and still feel personal. And regime dependence: below roughly a $20 gold daily range, my system mostly stands aside, because there is not enough movement to pay for the losses. In the release 1.26 backtest the maximum drawdown was 11.2 percent. Holding through that is the actual price of the method.

If you need a high hit rate to stay calm, this exit style will be miserable for you, and that is worth knowing before you choose it rather than after.

How did I almost fool myself while testing this?

Twice, both times silently.

The first sweep looked like a perfect null result: every variant identical to the baseline. It turned out the setting was read as a percentage and I had passed it as a fraction, so I was banking a fraction of one percent instead of fifteen to fifty. The second trap was that my usual per-trade metric, measured in risk units, cannot see a partial close at all. It reported the same total for every variant while the actual outcome moved a lot.

Both are now caught automatically. Any tested setting that never moves the score gets flagged as unwired or invisible, never reported as "no effect". A null result is only a result if the test could have shown something else.

Frequently asked questions

Is it better to take profit early or let winners run? It depends on the shape of the strategy. On a breakout system where a few large winners carry the result, letting the full position run tested better on every variant I tried. On my gold systems one full target hit pays for about five full losses.

Does taking partial profits reduce drawdown? Not in my testing. Banking a share of the position early left the maximum drawdown deeper than holding, not shallower, because it shrank the big winners that offset losing runs while leaving full-size losers untouched. The result fell at every banked share I tried.

What is the disposition effect in trading? It is the tendency to close winning trades too soon and hold losing trades too long. Realising a gain feels good and realising a loss feels like an admission. A written exit rule moves that decision out of the moment where the bias operates.

Why do a few trades produce most of the profit in a breakout system? Because most breakouts fail or stall, and the system is built to lose small on those and stay in the few that keep going. On my five-minute gold system the best 30 of 1,643 backtest trades made up about three quarters of the result.

Should I move my stop to break-even? That is your call and depends on your system. Mine moves the stop to break-even after a trade has moved far enough in its favour and then trails it, which turns many would-be losses into scratches. The trade-off is that some trades get scratched just before they would have run.

How many exit rules should you test before trusting one? Enough that the winner is not just the luckiest of the batch, and with the decision rule written before you look. I tested more than twenty alternatives against my existing exit. None beat it, which is itself the useful result.

Does a trailing stop give back profit? Yes, every time. It only knows a move is over after price has turned, so some of each big winner is handed back. In my testing that give-back was cheaper than every rule that tried to exit sooner.

If you want to see where this ended up, MasterVP is my own intraday method built on public volume profile theory. It ships as the Master Volume Profiler indicator on TradingView and MetaTrader 5, which shows the read and leaves the order with you, and as Master Volume Sniper on MQL5 market, an Expert Advisor that executes the rules and risk settings you configure. Where do you take your exits from, a rule written before the trade or a feeling about the open profit?

---

This article was composed from the exits posts of the MasterVP build log series. Educational purpose only. Not financial advice. Figures cited are backtest and research results on real tick data, and past performance does not guarantee future results. Every trader remains responsible for their own risk settings and decisions.