Reward-to-risk is geometry, not a win-rate forecast

A reward-to-risk ratio compares a proposed favorable price distance with an adverse distance. It does not state the probability of reaching either level. A 2:1 ratio means the target distance is twice the stop distance under your chosen references; it does not mean a two-thirds chance of winning.

When reviewing ChartPilot’s levels, review current signup options and compare the explanation with the image. Use the arithmetic below to understand what the ratio measures and what information is still missing.

Access checked October 26, 2026: New accounts use a seven-day Premium trial with payment-method setup, then $59 USD/month unless canceled before trial end; applicable taxes appear at checkout. There is no ongoing free plan for new accounts. Existing free accounts retain their allowance. Review the current terms.

Give the ratio an explicit direction

People sometimes write risk-to-reward and sometimes reward-to-risk. State which quantity is first. This article uses reward-to-risk: target distance divided by stop distance. For an invented long scenario with entry 200, stop 195 and target 210, the distances are 10 and 5, so the ratio is 2:1.

Those three prices contain no count of successful or unsuccessful outcomes. They also do not identify an order type, fill price or trading cost. The reference-level consistency guide covers the geometry; this article focuses on the separate question of probability.

The same ratio, two different fictional records

Assume a deliberately simplified model: every winning outcome earns exactly two units, every losing outcome loses one unit, and there are no costs, partial exits, gaps or missed fills. A unit here is just a common measuring scale. It is not an amount you should risk.

In fictional set A, four of ten outcomes win and six lose. The total is four times two minus six times one: positive two units. In fictional set B, two win and eight lose. The total is four minus eight: negative four units. Both sets use exactly the same 2:1 payoff assumption.

The difference comes from the outcomes, not from changing the ratio. Neither set estimates the probability of a real ChartPilot assessment. They simply demonstrate why a target-to-stop distance alone cannot establish expected performance.

CME’s mathematical-expectation lesson combines outcome frequency with average win and loss size. That combination is the important distinction: knowing only the size relationship leaves the frequency unknown.

Break-even arithmetic is conditional

Under the same two-outcome, no-cost model, let p be the probability of the winning outcome. Expected units are 2p minus 1 times (1 minus p), or 3p minus 1. Setting this expression to zero gives p equal to one third.

This says that a one-third win probability would break even in that model. It does not say that a proposed 2:1 setup actually has that probability. Confusing a required threshold with an estimated probability turns an algebra exercise into an unsupported forecast.

If each modeled outcome also costs 0.1 unit, expected units become 3p minus 1.1. The zero point becomes about 36.7%. This cost is invented for illustration. Real costs and realized payoffs can vary, so do not treat this number as a universal threshold for your instrument or strategy.

Planned distances and realized payoffs are different records

The 2:1 drawn before a review is a planned relationship. A later execution record may contain different entry prices, partial exits, fees or a stop filled away from its reference. Those differences can change the realized average win and loss even if every original screenshot displayed 2:1.

A screenshot with a later high above the target is not sufficient to reconstruct execution. You may not know whether the entry happened first, whether an order was active, or whether a stop was encountered before the target within the same candle. Keep unknown sequences unknown.

If you record outcomes for study, define the rules before classifying the charts. Preserve ambiguous cases rather than dropping them because they make a summary less attractive. A small selected sample can describe itself; it cannot establish a reliable probability for every future chart.

A farther target changes the question

Moving the invented target from 210 to 215 raises the drawn ratio from 2:1 to 3:1 while leaving entry and stop unchanged. It does not add evidence that price can travel the extra distance. Inspect what lies between the entry and the target, including visible opposing structure.

Likewise, moving a stop closer improves the numerical ratio while changing the condition under which the scenario would be invalidated. Do not adjust reference levels solely to make the ratio look favorable. First explain why each level belongs in the chart interpretation.

Read confidence and targets on their own terms

The ChartPilot terms describe Low, Moderate and High confidence as judgments about visible evidence, not a measured win probability. A calculated risk-projection target is labeled and is not observed support or resistance. Neither label supplies the missing p in the example above.

Ask three questions when reading a report: Do the distances calculate correctly? What visible evidence supports the scenario? What remains unknown about probability and execution? A clear answer to the first question does not automatically answer the other two.

Review current signup options to use a chart assessment as a structured second opinion. Keep the ratio attached to its stated reference levels, and keep any claim about success rates attached to actual, appropriately collected evidence.