The InDecision Framework: How I Built an 82.5% Accurate Market Predictor
Seven years of trading data. Six factors. One framework that removes emotion from the equation. Here's how InDecision works — and why it outperforms gut instinct every time.
Seven years of trading data. Hundreds of patterns. One repeating truth:
Markets are predictable — if you know what to measure.
Signal: InDecision Framework accuracy stands at 82.5% over 7+ years of live trading data. Not backtested theory — real capital, real results.
Why Most Traders Fail
Most retail traders fail for one reason: they confuse conviction with emotion.
They see a chart moving up and feel bullish. They watch Bitcoin dump 10% and panic-sell. Their framework is their feelings — and feelings are the worst possible trading system.
I built InDecision to solve this. A cold, mathematical, multi-factor scoring model that delivers a single output: directional bias with a conviction percentage.
The Six Factors
Factor 1: Daily Pattern (30%)
The highest-weighted factor. Every asset has a statistically significant intraday pattern — times when buying pressure consistently dominates, and times when selling pressure dominates. InDecision maps these patterns with 30-minute resolution.
Factor 2: Volume Profile (25%)
Volume doesn't lie. Abnormal volume spikes signal institutional activity. Below-average volume signals distribution. InDecision scores volume against 14-day and 90-day averages across multiple exchanges.
Factor 3: Timeframe Alignment (20%)
A 15-minute bullish signal against a daily bearish trend is noise. InDecision checks alignment across 5 timeframes — from 15m to weekly — and scores signal coherence.
Factor 4: Technical Confluence (15%)
Key levels: Support/resistance, VWAP, Fibonacci retracements, Bollinger bands. When multiple technical structures align, conviction increases. InDecision weights the density of confluence.
Factor 5: Market Timing (10%)
Macro context matters. InDecision incorporates market session timing (NY open, London open, Asia), macro calendar events, and correlation with BTC dominance.
Factor 6: Risk Context
Not scored individually — applied as a multiplier and gate. If risk context is extreme (e.g., pre-Fed announcement), conviction is capped regardless of signal strength.
The Output
InDecision delivers:
BIAS: BULLISH
CONVICTION: 74.3%
FACTORS: 5/6 aligned
CAUTION: Volume below 90d avg
Insight: A conviction above 65% is my threshold for entering a position. Below that, I wait. Patience is a factor InDecision can't teach — that's on the trader.
What InDecision Is Not
It is not a trading bot. It does not execute. It does not replace judgment.
It is a signal infrastructure — the analytical layer that tells you whether to take a trade, not which trade to take.
The difference matters.
Subscribe to the newsletter for weekly InDecision Framework signals and analysis.
Explore the Invictus Labs Ecosystem
Follow the Signal
If this was useful, follow along. Daily intelligence across AI, crypto, and strategy — before the mainstream catches on.
From Framework to Signal: Building the InDecision API
The InDecision Framework ran for 7 years as a closed system — Python scorers feeding Discord and a trading bot. Turning it into a public API forced architectural decisions that changed how I think about signal infrastructure.
The Signals Were Real: InDecision Framework Hits 93% Win Rate in Live Markets
The bot was cycling every 2 minutes — its own watchdog killing it every 129 seconds. The signals inside were perfect: 86–100/100, 92% accuracy, calling direction while the market priced uncertainty at 50/50. One coding session fixed the infrastructure. The rest is on-chain.
The Trader Behind the Bot
PolyEdge scored 65/100 and sat. Knox traded manually and went 3 for 3. The gap between those two outcomes reveals the exact mechanisms the model was missing.