Why Fixed-Percentage Targets Fail
Fixed-percentage targets produce levels that are unreachable on low-volatility assets and hit by noise on high-volatility ones. Sizing every level from the asset’s Average True Range fixes this, and deriving the target at twice the stop distance holds reward-to-risk constant so the win rate stays interpretable.
A fixed percentage target is the most common mistake in retail signal systems. “Take profit at 3%” sounds disciplined, but 3% is a rounding error for Solana and a major move for EUR/USD. The same number produces targets that are unreachable on one asset and hit by noise on another.
Size the band to the asset, not the account
The fix is to express every level in units of the asset's own recent volatility, using Average True Range (ATR-14). The structure is then identical for every asset, and it is the ATR that differs:
| Level | Distance from entry | Why |
|---|---|---|
| Entry zone | ± 0.25 ATR | a band a reader can fill inside, not a single stale price |
| Stop | 2.5 ATR | beyond ordinary noise, close enough to keep the target reachable |
| First target | 3.0 ATR | twice the stop distance — a fixed 2:1 reward-to-risk |
| Second target | 5.25 ATR | a stretch level for scaling out; not scored |
Bitcoin's ATR is far larger than Apple's, so its stop is far wider in dollars — and exactly the same in risk terms. A stop tight enough to be sensible on Apple would be triggered by an ordinary day of Bitcoin trading.
Why true range, and not closing prices
Volatility can also be estimated from closing prices alone, and it is tempting because the data is simpler. It has one serious blind spot: it cannot see a gap. A stock that closes at 100 and opens at 94 has moved 6%, and a close-to-close estimate records that as a single ordinary change while a trader is already through their stop.
True range is the greatest of the bar's own high-low span, the distance from the previous close to the high, and the distance from the previous close to the low — so the gap is inside the measurement. That matters most for equities, which gap overnight and at weekends, and it is why sizing moved onto ATR.
The stop is derived, never chosen
The stop is always placed at exactly half the target distance, giving a fixed 2:1 reward-to-risk on every signal, on every asset, in every market condition.
Holding that ratio constant is what makes the track record legible. If reward-to-risk varied per signal, a win rate would be uninterpretable — you could not tell whether a good month came from better calls or from quietly widening the stops. With it fixed, break-even sits at a 33% hit rate and the win rate becomes a real measurement. That argument is laid out in why a 40% win rate can beat a 90% one.
On tuning
The multiples above are deliberately plain — one stop distance, one ratio, applied to every asset. An earlier version of this system instead fitted a separate multiplier for each asset class and timeframe against a backtest, which produced a table of six tuned numbers and a strong temptation to keep adjusting them. Fewer knobs is not a stylistic preference here: every parameter fitted to past data is a chance to describe that data rather than the market.
Where measurement genuinely earns its place is in deciding which signals are worth acting on rather than how wide to draw the levels. That test is run walk-forward — candidate rules are ranked on one half of the history and validated on a half withheld from the ranking — and it found that entries within 2 ATRs of the 50-period average substantially outperform entries allowed to run out to 4, with the same ordering in both halves. That is the filter behind the trade ideas we actually send.
How the resulting signals are then scored is covered in how a trading signal should be scored, and the live outcomes are on the track record.
Frequently asked questions
- Why do fixed-percentage take-profit targets fail?
- Because the same percentage means different things on different assets. A 3% target is a rounding error on Solana and a major move on EUR/USD, so one fixed number produces targets that are unreachable on low-volatility assets and hit by ordinary noise on high-volatility ones.
- How should a take-profit target be sized instead?
- In units of the asset’s own Average True Range (ATR-14). The stop sits 2.5 ATR from entry and the first target at twice that distance, so the structure is identical on every asset and it is the ATR that differs — which is exactly the point, since ATR is what "how much this thing moves" means.
- Why use Average True Range rather than closing prices?
- Because a volatility estimate built from closing prices cannot see a gap. A stock that closes at 100 and opens at 94 has moved 6%, but close-to-close records one ordinary change while a trader is already through their stop. True range is the greatest of the bar’s high-low span and its two distances from the previous close, so the gap is inside the measurement.
- How is the stop-loss placed?
- It is derived rather than chosen: 2.5 ATR from entry, with the first target at twice that distance, giving a fixed 2:1 reward-to-risk on every signal, on every asset, in every market condition. Holding that ratio constant is what makes the published win rate interpretable — at 2:1 you need to be right more than 33.3% of the time to break even before costs.
Related reading
Educational content, not financial advice. Trading carries risk, including loss of capital.