Why capital preservation comes first
The account has to still exist when the conditions you trade well finally appear.
9 min read · beginner · lesson 8 of 8
What this covers
- State why survival ranks above return for an account that compounds
- Explain what participation costs when conditions do not suit the method
- Recognise the account states in which no position is the correct position
- Separate opportunity cost from capital cost
- Set the limits that keep a single event from ending the account
The account is the thing being managed
A method produces results over a sample. Any single position is close to noise; the sample is where the method lives. This has one hard consequence: the account has to survive long enough to produce the sample.
That reframes what a loss is. Losing an amount inside the risk policy costs money. Losing an amount that forces the size down permanently, or ends participation, costs every future result the method would have produced.
Capital preservation is the recognition that compounding only happens to accounts that are still there.
Most conditions do not suit most methods
Every method has conditions it handles well and conditions it does not. Trend continuation performs badly in a range. Reversion performs badly in a strong trend.
Markets do not distribute those conditions evenly. A method can face weeks in which nothing it does well is on offer.
Participating anyway is where accounts drain without a single dramatic loss. No large error to point at. Just a long series of small ones taken in conditions the method was never built for.
What ordinary participation costs
IllustrativeAll figures are illustrative.
An account of 10,000 takes 40 positions in a month at 0.5 percent risk, in conditions the method handles poorly. Half reach target at 1R, half hit the stop.
Twenty gains of 50 and twenty losses of 50 is flat before costs. After spread and commission at, say, 6 per position, the month costs 240, or 2.4 percent.
Nothing went wrong. No rule was broken. The account is down 2.4 percent for having been present.
The same account taking eight positions in the conditions it handles well pays 48 in costs. Selectivity is about the cost of being present as much as about which positions work.
No position is a position
Holding nothing has a defined outcome. The account does not change, which beats participating in conditions where there is no edge to participate on.
No position produces no feedback and no sense of progress, so it is rarely the choice that feels like working. Make it a written rule.
Opportunity cost against capital cost
| Missing a move | Losing capital |
|---|---|
| Costs a result that was never yours | Costs results the account can no longer produce |
| Fully recoverable on the next opportunity | Recoverable only at the rate the table demands |
| No effect on future position size | Reduces every future position size |
| Feels worse | Is worse |
The limits that keep one event from ending it
Risk per position
A fixed percentage, applied uniformly. This bounds the ordinary loss.
Exposure to a single driver
Correlated positions counted as one. This bounds the loss when several positions turn out to be the same position.
Total open risk
A ceiling across everything held at once, so a portfolio of unrelated positions still cannot exceed a known figure.
A stop for the account, not just the position
A drawdown depth at which trading pauses. Without it, the other three limits still permit an indefinite slide.
The limits are bounds, not guarantees
A stop-loss is an instruction to exit at a level, not a promise of that level. Price can move past it without trading there. The fill arrives wherever liquidity resumes, which on a bad morning is a long way below where you asked to be out.
Weekend gaps, scheduled data and single-name events all do this. Size is the defence: a position small enough that a gap through the stop is survivable rather than defining.
Risk per position sits below what the arithmetic alone allows. The margin is there for the cases where the exit does not work as intended, and those are not rare enough to leave out of the sizing.
Key takeaways
- A method only exists over a sample, and the sample requires the account to survive.
- Participating in unsuitable conditions drains an account without a single visible error.
- Holding nothing has a defined outcome and is often the correct one.
- A missed move costs a result that was never yours; lost capital costs every future result.
- Four limits bound the damage: per position, per driver, total open, and account drawdown.
- Stops define intent, not the fill. Size is what makes a gap survivable.
Common mistakes
- Treating time in the market as evidence of working.
- Judging a flat month as a failure when the conditions did not suit the method.
- Setting per-position risk without any ceiling on total open risk.
- Assuming a stop-loss caps the loss at exactly the stop level.
Knowledge check
OptionalRelated
- Position sizing and risk per tradeMarket Foundations
- Drawdown and recoveryMarket Foundations