Reward-to-risk, decided before you enter
In short
A reward-to-risk ratio only means something if entry, target and stop were all set before the position existed. Afterwards, the stop moves because the loss feels large and the target moves because the gain feels sufficient. The stop belongs where the reason for the trade stops being true; position size is the variable that reconciles that placement with a fixed acceptable loss.
- Published
- Last reviewed
- Reading time
- 9 min
- Written by
- Barion AI
Risk defined afterwards was never defined
The most common failure in discretionary trading is not picking bad setups. It is picking reasonable setups and deciding the exit levels after the position is open.
Once you hold a position, every level you choose is contaminated. The stop moves because the loss feels large. The target moves because the gain feels good enough. Neither adjustment is analysis; both are the position arguing with you.
A reward-to-risk ratio has meaning only if entry, target and stop were all set before the position existed. Computed afterwards, it is a description of what happened rather than a constraint on what you were willing to accept.
The arithmetic, and what it hides
For a long position:
- Risk = entry − stop
- Reward = target − entry
- Ratio = reward ÷ risk
A setup entered at 100 with a stop at 96 and a target at 112 risks 4 to make 12: a 3:1. Straightforward.
What the ratio hides is that it says nothing about likelihood. A 10:1 setup that resolves favourably one time in twenty is a losing proposition. A 2:1 that resolves half the time is a good one.
Ratio and hit rate have to be considered together, which is why a published ratio floor is a filter, not a strategy. It removes setups whose geometry cannot work. It does not tell you the remaining ones will.
Why a floor at all
The case for refusing to publish anything below a threshold is not that low ratios always lose. It is about what a floor does to behaviour over many decisions.
| Ratio | Break-even hit rate needed |
|---|---|
| 1:1 | 50% |
| 1.5:1 | 40% |
| 2:1 | 33% |
| 3:1 | 25% |
| 4:1 | 20% |
Before costs. Spread, slippage and commission all move these upward.
At 1:1 you must be right more than half the time simply to stand still, and that is before any friction. At 2:1 you can be wrong twice as often as you are right and remain flat. The floor buys tolerance for being wrong, which is the only reliable assumption in this domain.
PSX Invest applies a minimum 2:1 on anything published. A setup that does not clear it is not ranked poorly — it is not published.
Where the stop actually belongs
The frequent error is placing the stop at a comfortable loss rather than at the level that invalidates the idea.
The stop belongs where the reason for the trade stops being true. If the setup rests on a support level holding, the stop sits below that level. If it rests on a trend continuing, it sits where the trend would be broken. Those placements are determined by the chart, not by the account.
Placing a stop at a round percentage of capital inverts the logic: it exits at a price that means nothing to the market, frequently just before the level that would have mattered.
Position size is what reconciles the two, which is the next section.
Sizing is the free variable
Traders often treat position size as fixed and stop distance as adjustable. It should be the reverse.
- Find the invalidation level. Chart-determined.
- Measure the distance from entry to that level. This is risk per unit.
- Decide what you are willing to lose on this idea, as a fixed fraction of capital, set in advance and applied uniformly.
- Divide. Size = acceptable loss ÷ risk per unit.
Wide stop, smaller size. Tight stop, larger size. The amount at risk stays constant across every position, which is what makes results comparable and prevents a single wide-stop idea from carrying disproportionate weight.
This also removes the most common rationalisation in trading — moving a stop closer than the analysis supports in order to justify a larger position.
Volatility changes the geometry
The same instrument needs different stop distances in different conditions. A level that is a meaningful break in a quiet market is ordinary noise in a volatile one.
This is what a volatility measure such as ATR is for. It does not indicate direction; it indicates how much room a position needs to avoid being stopped out by normal movement. Rising volatility means wider stops and, by the sizing rule above, smaller positions for the same risk.
Ignoring this produces a specific and frustrating pattern: correct directional calls stopped out by ordinary noise, followed by the move continuing without you.
Trailing, and when it helps
Moving a stop toward break-even after a position has advanced reduces exposure without capping the upside. PSX Invest publishes a rule of this kind — stops trail to break-even at the halfway mark toward target.
The value of a published rule is that it is decided in advance. A trailing rule you invent while watching a position is not risk management, it is the same improvisation the plan was supposed to prevent, arriving in a more respectable costume.
Two honest caveats: trailing reduces average hold time and will convert some eventual winners into scratches. It trades expectancy for variance reduction, and whether that is worthwhile depends on the strategy rather than being universally good.
The systematic version
The reason this is worth encoding in a system rather than leaving to discipline is that discipline degrades under exactly the conditions where it matters most.
A published call that carries an entry range, a target and a stop, all set before the position exists and all visible with the indicators behind them, removes the opportunity to redefine risk after the fact. The rule is applied identically on a calm Tuesday and a violent Thursday.
That is the same principle applied throughout Barion's work: a constraint enforced by the system is a boundary, and a constraint enforced by intention is a request.