1
Hypothesis
State why a repeatable market behavior may produce an edge.
Lesson · Backtesting and Trading Systems
Testable Trading Rules and Hypotheses
Convert a market idea into deterministic entry, exit, sizing, and invalidation rules that produce the same decision from the same data.
Core concepts
What to understand
Keep each definition operational so it can be implemented and tested.
2
Deterministic rule
The same inputs must always produce the same signal.
3
Complete lifecycle
Define no-trade, entry, sizing, management, exit, and invalidation conditions.
Interactive lab
Trading Rule Builder
Build a long or short rule, inspect its generated signals, and verify that the rule is complete enough to test.
Trading Rule Builder
Deterministic synthetic educational data · no live market connection
BACKTEST MODEL
Process
Convert observations into conditions
Use exact values and event timing.
Observe
Describe the price and volume behavior.
Specify
Translate it into measurable conditions.
Reject ambiguity
Remove discretionary words that cannot be calculated.
Knowledge check
Check your understanding
Select one answer for each question. Feedback appears immediately.
1. Which statement is directly testable?
The breakout rule specifies data, threshold, and execution timing.
2. What should a complete rule include?
A system needs the complete trade lifecycle and accounting rules.
3. Why define the hypothesis first?
The hypothesis connects the rule to a specific market behavior.
4. A rule reads: buy when the close exceeds the prior 20-bar high. Today's close is 104.20 and the prior 20-bar high is 104.00. The earliest price the rule can fill at is:
The signal is known at the close of the bar, so the next executable price is the following open — the signalTime and execution fields in the lab's exported rule say exactly this.
Practical rules
- Use exact thresholds and time references.
- Specify when the signal is known and when the order is placed.
- Define conditions that prevent a trade.
Common mistakes
- Using words such as strong, weak, or near without values.
- Testing entry rules without exits.
- Allowing manual exceptions during the test.
References
Official reading
The labs are simplified simulations. These sources cover historical performance, data, costs, or execution limitations.
FINRA · Smart Beta—What You Need to Know
Explains that historical backtesting is hypothetical and does not predict future performance.
FINRA Rule 2210 · Communications with the Public
Includes restrictions on performance projections and misleading performance claims.