You can have a decent entry and still bleed a funded account dry — because how you manage the trade after the fill decides most of the result. The problem is that almost nobody actually measures whether their management is helping or quietly costing them money.
The uncomfortable starting point
Most traders lose. Not because their entries are garbage, and not because they picked the “wrong” strategy off YouTube. They lose because they never operate like a business: no risk discipline, no record of what they actually do, and no idea whether their exits add or subtract from their edge. The entry gets 90% of the obsession and maybe 10% of the impact.
Trade management — scaling out, trailing a stop, holding to a fixed target, or taking a flat R — is where funded accounts are won and lost. And here’s the honest part almost no course will tell you: there is no single correct method. Scaling out smooths equity but caps your winners. Holding to target maximizes big runners but stomachs more full stop-outs. Trailing captures trends but chops you up in ranges. One approach fits one system; the opposite fits another. The only way to know which fits yours is your own numbers, over a real sample.
Define “working” before you measure it
“Working” isn’t a feeling. Pick concrete definitions up front, or you’ll rationalize whatever you did last:
- Higher expectancy per trade — average R won across the whole sample, winners and losers combined.
- A smoother equity curve — smaller peak-to-trough drawdowns, which matters enormously on a trailing drawdown account where the floor chases your highs.
- Fewer rule breaches — management that keeps you inside daily loss and consistency limits, not just theoretically profitable.
A method that lifts expectancy but triples your drawdown might still fail you under a funded account’s constraints. Both axes matter. Run your candidate exits through an expectancy calculator so you’re comparing a single honest number, not a highlight reel of your best day.
The measurement most traders skip: R-multiples
You cannot compare exits in dollars — position sizes drift, instruments differ, volatility changes. Normalize everything to R, where 1R is the amount you risked on that trade. A winner that pays 2.3× your risk is +2.3R; a full stop-out is −1R. Now every trade speaks the same language and you can actually add them up. If R is new to you, start with the R-multiple primer and the trading expectancy breakdown.
Once your trades are in R, you’re not arguing about “the one that got away.” You’re reading a distribution.
Run the honest experiment
Here’s the method that separates measurement from myth:
- Log every trade the same way — entry, stop, the exits you actually took, and the exit each candidate rule would have produced (e.g. “held to 2R target” vs “trailed by structure”).
- Tag the setup and the rule. You’re not evaluating “your trading” in a blob — you’re evaluating this setup managed this way.
- Collect a real sample. A handful of trades is noise. You need enough that the result isn’t luck — think dozens per approach before you trust it, more if your win rate is streaky.
- Compare expectancy AND drawdown, side by side.
| Management style | Tends to raise | Tends to cost | Fits systems that… |
|---|---|---|---|
| Scale out partials | Win rate, equity smoothness | Average winner size | Have frequent, modest moves |
| Hold to fixed target | Average R per winner | Win rate, comfort | Produce clean, measured runs |
| Trail the stop | Big-trend capture | Consistency in chop | Ride sustained momentum |
| Flat fixed R | Simplicity, sample clarity | Upside on outliers | Are still being validated |
The table shows tendencies, not verdicts. Your data decides which row is true for you.
Sample size is the thing that lies to you
The single most common self-sabotage: switching methods after five trades. Five trades can’t tell scaling out from holding — the variance swamps the signal. You need a large-enough run that the difference is unlikely to be chance. This is exactly where a Wilson confidence interval earns its keep: it tells you not just the win rate, but how sure you’re allowed to be given how many trades you’ve logged. A 70% win rate over 10 trades and over 100 trades are wildly different claims.
See it in Shibiki
This is the part that’s brutal to do by hand and easy to fake in a spreadsheet. In Shibiki, you’d see an edge-health panel for each setup: expectancy in R, the win rate with its Wilson confidence band, and an R-multiple distribution showing where your results actually cluster. Tag two management rules and you’d get a side-by-side — the same setup, held-to-target versus trailed — each with its own expectancy and confidence, plus the equity curve each would have carved out. Because every trade is auto-journaled, you’re comparing real fills, not memory. You’d stop debating exits and start reading them. (No invented numbers here — the panels fill from your trades.)
That’s the whole point: professionals don’t have magic entries. They have a measured edge and management they’ve proven fits it.
Protect the account while you test
Running the experiment doesn’t mean risking the account to find out. Keep position size mechanical with a position size calculator, pre-check your reward-to-risk with the risk-reward calculator, and stay honest about how a bad stretch dents a trailing drawdown floor. Test on live trades if you must, but never on live recklessness.
Measure it. Keep what your numbers reward. Drop what your ego is defending.
Related: Expectancy calculator · R-multiple explained · Trailing drawdown