Salidas parciales: ¿tomar la mitad en 1R o dejar correr toda la operación?
Quick answer: there is no universal answer, but there is one number that decides it. If many of your trades reach 1R and then reverse, taking a partial improves your expectancy. If most of the trades that reach 1R go on to target, the partial reduces it. The break-even point usually sits near one in four.
Puntos clave
- A partial does not reduce the risk of the trade: you took that on at entry.
- It lowers variance and clips your good tail at the same time.
- The decision comes from your history, not from a preference.
- If you take partials out of nerves, the problem is your size, not your method.
How each option behaves
Assume 1R of risk and a 2R target. The two ways to manage it:
| Qué ocurre | All to target | Half at 1R + stop to breakeven |
|---|---|---|
| Never reaches 1R | −1.0R | −1.0R |
| Reaches 1R, then reverses | −1.0R | +0.5R |
| Reaches the 2R target | +2.0R | +1.5R |
That is the whole trade-off: partials turn losers into small winners, and big winners into medium ones.
The table that decides it
Take 100 trades where 40 never reach 1R. The only thing that changes is how the other 60 split:
| Of the 60 that reach 1R | All to target | Partial at 1R | Winner |
|---|---|---|---|
| 30 reverse / 30 run on | −0.10R | +0.20R | Parcial |
| 20 reverse / 40 run on | +0.20R | +0.30R | Parcial |
| 15 reverse / 45 run on | +0.35R | +0.35R | Tie |
| 10 reverse / 50 run on | +0.50R | +0.40R | Let it run |
| 0 reverse / 60 run on | +0.80R | +0.50R | Let it run |
The crossover sits at 15 out of 60 — one in four. Above that reversal rate, partials win; below it, they cost you. This is not a matter of opinion. It is a matter of measuring your own rate.
How to measure your number
- Take your last 50 closed trades.
- Mark which ones reached 1R in your favour at any point.
- Of those, count how many finished as losses.
- Divide: that fraction is your reversal rate.
- Above about 25%, partials probably suit you.
One caveat: 50 trades is the minimum for the number to mean anything, and it is worth recalculating periodically. A high reversal rate may also be telling you that a 2R target is too ambitious for the instrument or session you trade.
Errores comunes
- Taking the partial at a different point each time, depending on how you feel.
- Moving the stop to breakeven too early and getting taken out by normal noise.
- Taking a partial and then re-entering, paying costs twice.
- Using partials to justify a position size you cannot tolerate.
- Comparing the two methods over twenty trades and drawing conclusions.
How to keep it fixed
Whichever you choose, what ruins the result is changing it mid-trade. The Cortex Signal Kit delivers signals with the target defined from the start, and Cortex Automation applies the same management every time without your mood getting a vote. To measure your reversal rate with an organised log, there is the free ebook The Trading Journal Blueprint.
So should I take partials or not?
It depends on your reversal rate. Above roughly one in four trades reaching 1R and turning around, partials usually improve expectancy.
Where should the partial go?
1R is the natural point because it leaves the trade with no net risk. Taking partials before 1R rarely pays.
Should I move the stop to breakeven when I take a partial?
It is common, but it has a cost: you increase exits caused by noise. Move it only once price has pulled away from the level.
Is exiting in three parts better?
It adds complexity and cost without changing much. Two parts already capture most of the effect.
Does this apply to an automated system?
Yes, and it is easier to measure there: you can test both management styles over the same history and compare expectancy.
Decide once, execute it the same way every time. See the Cortex Signal Kit or try it with the Automation Test.
Disclaimer: educational content. Numerical examples are illustrative and should be validated with your own data. Cortex Next is not a financial service and guarantees no results. Trading involves risk of loss.
