Synthetic Indices Strategy Guide — Analysis scheme pro
Deriv synthetic indices (such as the Volatility Indices) are generated by a cryptographically secure random number generator, so no tool can predict the next tick with certainty. What a trader can do is verify whether recent behaviour matches the theoretical distribution, measure how strong any deviation is, and confirm it across multiple window sizes before acting. This guide explains how to use each part of the terminal for that workflow.
1. Start with the digit heatmap
Every digit contract (Matches, Differs, Over, Under, Even, Odd) is settled on the last digit of the price quote. The Digit Distribution page shows how often each digit 0–9 has appeared in your chosen window, against the 10% baseline you would expect from a uniform generator.
A digit showing as HOT or COLD is a historical observation, not a signal that it is "due" to continue or revert. Use it as the starting point for further checks, never as a decision on its own.
2. Confirm across multiple windows
A deviation visible in a 25-tick window can be pure noise. The multi-window comparison (MICRO 25 / SHORT 50 / MEDIUM 100 / LARGE 250–500 / EXTENDED 1000) recalculates the same statistics at each scale. Treat a pattern as interesting only when it points the same way in several windows; the terminal labels single-tiny-window patterns as "SHORT-TERM PATTERN — LOW RELIABILITY" and strong disagreement as "NO ROBUST EDGE DETECTED".
3. Check the statistical engine, not just the percentages
The Insights page reports entropy, chi-square uniformity tests, z-scores and confidence intervals. A large percentage gap on a small sample can still be statistically insignificant — always read the sample size and the uncertainty range next to any percentage before trusting it. Statistical deviation is not predictive evidence.
4. Use the contract scanners with correct baselines
Matches wins at a 10% reference probability and Differs at 90%. Over/Under depends on the barrier you select, and a digit equal to the barrier wins neither side. The Matches/Differs and Over/Under scanners compare observed rates with these exact contract probabilities so you are never comparing against the wrong baseline.
5. Validate forward before trusting anything
Every prediction card the engine produces is logged with its probability, confidence and statistical snapshot, then automatically graded WIN or LOSS against the next tick. The History page shows whether the engine is actually beating its own baseline on live data. The Backtest Lab goes further: it splits historical ticks chronologically into train, validation and out-of-sample test sets so results are not inflated by tuning on the same data.
6. Respect the limits
Synthetic indices have no memory: each tick is independent, so streaks and imbalances arise randomly and can vanish at any moment. This terminal will sometimes report NO RELIABLE EDGE — that is a correct and useful answer. Nothing here is financial advice, no outcome is guaranteed, and version 1 places no trades. Never risk money you cannot afford to lose.
Risk notice
Statistical analysis does not guarantee future outcomes.
Past tick distributions do not guarantee the next digit.
Short-term digit imbalances may occur randomly.
Version 1 is analysis only — no trades are placed. No Martingale or loss-recovery staking is provided by design.