Trading signals, defined
Trading signals are written or programmed outputs intended to indicate a potential trading decision, such as entering a trade, exiting a trade, or changing the timing. In forex, they are often described as “alerts” generated by a person’s rules or by a method that processes market information. The key idea is that a signal is a decision prompt, not the trade itself.
A useful way to think about trading signals is as a translation layer: raw information (for example, price or technical measures) is converted into a discrete instruction-like output (for example, “enter” or “close”). This conversion depends on the signal’s rules, assumptions, and the specific inputs it uses.
How trading signals work in forex
Most trading signals follow a simple model:
- Inputs: The system or person uses defined data. This might include price series, derived indicators, volatility measures, or other features.
- Rules: A rule set decides when conditions meet a threshold. Examples include “if a condition is true, then signal X.” These rules may be manual (discretionary) or automatic (algorithmic).
- Output: The method emits a signal, typically linked to a time when the decision should be acted on.
- Execution: A separate step turns the decision into an order. Execution depends on liquidity, available spreads, order type, latency, and platform behavior.
It helps to separate stable mechanics from variable conditions. The stable mechanics are the input-to-output conversion and the decision timing. The variable parts are everything that affects what happens after the prompt is created: changing market conditions, costs, fills, and any delays between signal generation and order placement.
A simple example with stated assumptions
Assume a signal rule issues a “buy” prompt whenever a chosen measure crosses a threshold at a specific time. If the trader places an order at the next available moment, the realized result will depend on the price at execution time, not the price observed by the signal rule. If execution happens later or at a worse price, the outcome can differ substantially from what a backtest implied under idealized assumptions.
Limitations, risks, and failure modes
Trading signals can fail in several material ways:
- Assumption break: Rules often rely on patterns that may stop appearing. When the market regime changes, the same conditions can produce different results.
- Data and timing issues: If the signal uses delayed or different data than the trader sees, the signal can be based on information that is no longer current.
- Execution and costs: Even if a signal is “correct” according to its internal logic, transaction costs (spreads, commissions, slippage) can turn an expected move into an unfavorable outcome.
- Overfitting and limited history: A method can perform well in past samples while failing in new conditions, especially when tuned too closely to historical data.
- No risk controls by default: A signal may tell you what to do, but not how to manage position sizing, stop logic, or exposure. Without risk control, small errors can compound.
Because outcomes vary with costs, execution quality, and market behavior, historical relationships do not establish future results.
How to verify claims about signals (without guessing)
Independent verification is possible, but it must mirror real usage. A practical verification approach includes:
- Clarify the exact rule: What inputs are used, what thresholds trigger signals, and how often signals are generated.
- Define the action mapping: Does “enter” mean a market order at the next timestamp, or a limit order at a specified price?
- Model realistic execution: Incorporate assumptions about spreads, slippage, and the time gap between signal creation and order placement.
- Test across different periods: Use multiple time ranges to reduce the chance that results depend on one favorable regime.
- Check failure cases: Review periods when the signal underperforms to understand which assumptions broke.
If a signal cannot be verified from its stated rules and assumptions, its usefulness is harder to assess objectively.
What to look for in adjacent concepts
Trading signals are often discussed alongside related ideas:
- Indicators: Indicators are calculations on data. A signal is typically a decision output derived from indicators or other measures. - Automation: Automated execution is the process of placing orders based on rules. A signal can exist without automation.