Methodology & Risk Factor Breakdown
A transparent, data-driven approach to Bitcoin risk assessment using five independent pillars.
Overview
The Bitcoin Risk Dashboard provides a data-driven composite score (0–100) based on a weighted blend of independent pillars. Each pillar captures different aspects of market risk: liquidity conditions, momentum indicators, leverage metrics, social sentiment, and macro overlays. The system is updated daily with transparent, documented data sources and provides clear guidance through risk bands that translate scores into actionable insights.
What is the BTC G-Score?
The BTC G-Score is a composite risk assessment score ranging from 0 to 100, where:
- 0-14: Aggressive Buying - Maximum allocation recommended
- 15-34: Regular DCA Buying - Continue regular purchases
- 35-49: Moderate Buying - Reduce position size
- 50-64: Hold & Wait - Hold existing positions
- 65-79: Reduce Risk - Consider taking profits
- 80-100: High Risk - Significant risk of correction
Score Calculation
The G-Score is calculated by taking a weighted average of multiple risk factors across five pillars:
Pillar and factor weights are configurable and sum to 100%. Live defaults are shown below:
- Liquidity/Flows (35%): Defaults: Stablecoins 21%, ETF Flows 9%, Net Liquidity 5% (pillar total 35%)
- Momentum/Valuation (25%): Defaults: Trend & Valuation 20%, On-chain Activity 5% (pillar total 25%)
- Term Structure/Leverage (20%): Derivatives and funding rates
- Macro Overlay (10%): Macroeconomic conditions (DXY, 2Y rates, VIX). Net Liquidity appears here for context only; it is scored under Liquidity (5%) to avoid double-counting
- Social/Attention (10%): Social sentiment indicators
Note: On-chain Activity contributes to Momentum rather than standing alone. Net Liquidity is scored under Liquidity (5%) but also displayed in Macro for context without affecting the composite score.
BMSB-Led Trend Analysis: Inside Trend & Valuation, Distance to the Bull Market Support Band carries the largest weight by design (60%), reflecting where Bitcoin sits relative to its dynamic support levels. Long-trend stretch (Mayer Multiple, 30%) and weekly momentum (RSI, 10%) contribute the rest.
Adjustments
The base factor score may include small adjustments to account for market context:
CycleAdjustment
A gentle context nudge derived from Bitcoin's deviation from its long-term power-law trend. Calculates where current price sits relative to Bitcoin's historical growth pattern. Only activates when deviation exceeds 30% from the trend line. Small magnitude (capped at ±2.0 points).
SpikeAdjustment
A fast-path nudge when today's price move is a significant outlier versus recent volatility (20-day EWMA). Only activates when daily move exceeds 2x recent volatility (Z-score >2.0). Large up-spikes → small risk increase; large down-spikes → small risk decrease. Capped at ±1.5 points.
Note: Both adjustments are transparent and additive to the factor-blended composite. They are displayed as signed numbers with 1 decimal precision (e.g., +1.3, −0.7). If not present or zero, they show "—".
Risk Bands
Bands translate the 0–100 score into plain-English guidance. They're not trade signals, but rather risk assessment tools to help inform decision-making.
Risk Factor Breakdown
Trend Valuation
What we look at:
- •Bull Market Support Band (BMSB) distance - 60% weight
- •Price vs 200-day SMA (Mayer Multiple) - 30% weight
- •Weekly momentum (RSI-14 on weekly samples) - 10% weight
Why it matters:
BMSB-led approach captures cycle-aware trend analysis. BMSB (20W SMA / 21W EMA) provides dynamic support levels that adapt to market cycles, while Mayer Multiple adds valuation context and weekly RSI confirms momentum.
How it affects risk:
↑ BMSB distance + ↑ Mayer Multiple + ↑ weekly RSI ⇒ ↑ risk; price near/below BMSB with weak momentum ⇒ ↓ risk
Update cadence & staleness:
Daily updates; stale >6h
Primary sources:
Caveats:
True BMSB calculation using weekly resampling. Weekly RSI samples every 7th day. Requires 50+ weeks of data for full calculation.
Net Liquidity
What we look at:
- •Net Liquidity Level (WALCL - RRPONTSYD - WTREGEN) - 15% weight
- •4-week Rate of Change (short-term trend) - 40% weight
- •12-week Momentum/Acceleration (trend strength) - 45% weight
Why it matters:
Momentum-focused approach emphasizes directional changes over absolute levels. Rate of change captures short-term liquidity trends, while momentum shows acceleration/deceleration patterns that often precede market moves.
How it affects risk:
↑ net liquidity + ↑ growth rate + ↑ acceleration ⇒ ↓ risk; contracting liquidity momentum ⇒ ↑ risk
Update cadence & staleness:
Weekly updates; stale >10 days
Primary sources:
Caveats:
Macro proxy; RRP data may be sparse during normalization periods. Indirect relationship to BTC.
Stablecoins
What we look at:
- •Aggregate Supply Growth (USDT 65%, USDC 28%, DAI 7%) - 50% weight
- •Growth Momentum (7d vs 30d acceleration) - 30% weight
- •Market Concentration Risk (HHI diversification) - 20% weight
Why it matters:
Multi-stablecoin approach captures total crypto buying power. Supply growth indicates liquidity expansion, momentum shows sustainability, and concentration measures systemic risk from dominance.
How it affects risk:
↑ aggregate supply growth + ↑ momentum + ↓ concentration ⇒ ↓ risk; supply contractions or high concentration ⇒ ↑ risk
Update cadence & staleness:
Daily updates; stale >48h
Primary sources:
Caveats:
Exchange behavior, regulatory events, and chain migrations can create temporary distortions. HHI reflects current market structure risk.
Etf Flows
What we look at:
- •21-day Rolling Sum (all ETFs combined) - 30% weight
- •Flow Acceleration (7d recent vs previous 7d) - 30% weight
- •ETF Diversification (HHI concentration risk) - 40% weight
Why it matters:
Diversification-focused institutional demand analysis. Rolling sum captures sustained momentum, acceleration shows trend changes, and diversification measures systemic risk from single ETF dominance (most important factor).
How it affects risk:
↑ sustained inflows + ↑ acceleration + ↑ diversification ⇒ ↓ risk; concentrated outflows or single ETF dominance ⇒ ↑ risk
Update cadence & staleness:
Business days; stale >1 day
Primary sources:
Caveats:
Holidays/reporting lags. HHI reflects current ETF market structure. Schema changes may affect data parsing.
Term Leverage
What we look at:
- •Funding Rate Level (BitMEX 30-day average) - 40% weight
- •Funding Rate Volatility (instability measure) - 30% weight
- •Term Structure Stress (funding-spot divergence) - 30% weight
Why it matters:
Multi-dimensional leverage analysis captures both intensity and instability. Funding levels show leverage demand, volatility indicates market stress, and divergence measures term structure health.
How it affects risk:
↑ funding rates + ↑ volatility + ↑ stress divergence ⇒ ↑ risk; negative funding + low volatility ⇒ ↓ risk
Update cadence & staleness:
Daily updates; stale >24h
Primary sources:
Caveats:
Single venue dependency (BitMEX). Extreme events may cause API failures. Funding-spot correlation assumes efficient arbitrage.
Onchain
What we look at:
- •Network Congestion (transaction fees vs history) - 60% weight
- •Mempool Activity (7-day avg mempool size in MB) - 40% weight
- •NVT Ratio (Network Value to Transactions proxy) - disabled
- •Hash Rate Security (network security context) - informational only
Why it matters:
Focused on-chain congestion analysis. Transaction fees capture demand pressure and network stress, while mempool size shows pending transaction volume. NVT proxy was unreliable and has been disabled.
How it affects risk:
↑ transaction fees + ↑ mempool congestion ⇒ ↑ risk (network stress); low fees + small mempool ⇒ ↓ risk
Update cadence & staleness:
Daily updates; stale >3 days
Caveats:
Fee spikes may be event-driven (ordinals, congestion). Mempool data converted from bytes to MB. Hash rate provides context but not scored.
Social Interest
What we look at:
- •Google Trends Bitcoin interest (proxy via available data) - 70% weight
- •Fear & Greed Index sentiment - 30% weight
Why it matters:
Social sentiment analysis using available free APIs. Google Trends captures retail interest and search volume spikes, while Fear & Greed Index provides market sentiment context from multiple data sources.
How it affects risk:
↑ search interest + ↑ fear/greed extremes ⇒ ↑ risk; low interest + neutral sentiment ⇒ ↓ risk
Update cadence & staleness:
Daily updates; stale >1 day
Caveats:
Limited to available free APIs. No direct Google Trends or social media integration. Fear & Greed Index aggregates multiple sentiment sources.
Macro Overlay
What we look at:
- •Dollar Strength (DXY 20-day momentum) - 40% weight
- •2-Year Treasury Yield (20-day momentum) - 35% weight
- •VIX Risk Appetite (percentile level) - 25% weight
Why it matters:
Simplified macro environment analysis focuses on three key Bitcoin risk factors: dollar strength affects international flows, rising short-term rates compete with risk assets, and VIX spikes indicate flight-to-quality away from risk assets.
How it affects risk:
↑ dollar strength + ↑ rising rates + ↑ VIX fear ⇒ ↑ risk; weak dollar + falling rates + low VIX ⇒ ↓ risk
Update cadence & staleness:
Daily updates; stale >1 day
Primary sources:
Caveats:
VIX can be volatile during market stress. Dollar strength effects vary by global liquidity conditions. FRED data may have reporting delays.
Data Sources
| Provider | Metrics Used | Cadence | Notes |
|---|---|---|---|
| Coinbase | Spot price, daily candles, historical backfill | Real-time | Primary price source for all calculations (700+ days) |
| FRED (St. Louis Fed) | WALCL, RRPONTSYD, WTREGEN | Weekly | Macro liquidity indicators |
| CoinGecko | Stablecoin supply, market data | Daily | Stablecoin tracking |
| Farside | ETF flows, institutional data | Business days | ETF flow tracking |
| BitMEX | Funding rates, basis | Every 8h | Derivatives data |
| Blockchain.info | Fees, mempool, miner revenue | ~10 min | On-chain metrics |
| Google Trends | Search interest | Daily | Social sentiment |
| CBOE | VIX volatility index | Business days | Equity volatility |
Price History & Technical Indicators
Unified Price History System
Price history is maintained in a local CSV file with daily UTC close prices. The system fetches 700+ days of historical data from Coinbase's public API using chunked requests to handle their 300-record limit. Daily operations append recent Coinbase candles and deduplicate existing records.
Benefits: No external API keys required, reliable data source with 2+ years of history, consistent price calculations across all factors, and automatic daily updates.
Technical Indicators
Bull Market Support Band (BMSB)
Calculated using 20-week Simple Moving Average and 21-week Exponential Moving Average of weekly closes. The band represents key support levels during bull markets and is the primary component (60% weight) of the Trend & Valuation factor.
50-Week SMA Diagnostic
A display-only indicator that shows Bitcoin's position relative to its 50-week Simple Moving Average. This appears as a pill on the Trend & Valuation factor card:
- Above 50W SMA: Gray pill shows current SMA value (e.g., "Above 50W SMA ($99k)")
- Below 50W SMA: Amber warning pill after 2+ consecutive weeks (e.g., "Below 50W SMA (3+ weeks)")
Note: This diagnostic is educational only and does not affect the risk score calculation.
Other Technical Calculations
- Mayer Multiple: Current price divided by 200-day Simple Moving Average
- Weekly RSI: 14-period Relative Strength Index calculated on weekly closes
- Weekly Resampling: Daily prices converted to weekly using ISO week boundaries (Sunday close)
Freshness Rules
Each factor has specific staleness thresholds based on its data update frequency. When factors become stale, they are excluded from the composite score calculation and weights are re-normalized among the remaining fresh factors. This ensures the composite score only reflects current, reliable data.
Status indicators: Fresh (green) • Stale (yellow) • Excluded (gray)
Glossary
Trading Terms
- Funding
- Periodic payments between long and short positions in perpetual futures
- Basis/Contango
- Futures price above spot price; indicates bullish sentiment
- Backwardation
- Futures price below spot price; indicates bearish sentiment
- Mempool
- Queue of unconfirmed Bitcoin transactions waiting to be processed
Technical Terms
- Mayer Multiple
- Bitcoin price divided by its 200-day moving average
- Net Liquidity
- Federal Reserve balance sheet minus reverse repo and Treasury account
- RSI
- Relative Strength Index; momentum oscillator measuring speed of price changes
- VIX
- Volatility Index; measures expected volatility in S&P 500
ETF Flow Predictions
Our ETF Predictions system uses advanced machine learning models to forecast Bitcoin ETF flows, providing insights into institutional demand patterns and market sentiment.
Prediction Models
- ARIMA
- Time series forecasting using historical patterns
- LSTM Neural Network
- Deep learning for complex pattern recognition
- Random Forest
- Ensemble method combining multiple decision trees
- Ensemble Method
- Weighted combination of all models for optimal accuracy
Key Features
- • 7-day rolling forecasts with confidence intervals
- • Individual ETF performance predictions
- • Market share analysis and trend identification
- • Real-time data integration from ETF flow sources
- • Historical accuracy tracking (87%+ for 1-day predictions)
- • Dynamic confidence scoring based on data consistency
How It Works
The system analyzes recent ETF flow patterns, calculates trend momentum, and applies machine learning models to generate predictions. Confidence levels are dynamically adjusted based on data variance and historical accuracy. Predictions are updated in real-time as new data becomes available.
Frequently Asked Questions
Why did my score change overnight?
Scores update daily as new data becomes available. Factors like funding rates, ETF flows, and social sentiment can change significantly between updates, affecting the composite score.
What if a source is down?
If a data source is unavailable, that factor will be marked as excluded and won't contribute to the composite score. The system will continue to function with the remaining fresh factors.
Are bands trading advice?
No. Risk bands are educational tools for risk assessment, not trading signals. Always do your own research and consider your risk tolerance before making investment decisions.
How often do weights change?
Weights are configurable but typically remain stable. You can use the "Weights" tool to preview how different weightings would affect the composite score without changing the actual configuration.
Why does the 50W SMA pill always show?
The 50-week SMA diagnostic is always visible for educational transparency. When BTC is above the 50W SMA, it shows as a gray informational pill. When below for 2+ consecutive weeks, it becomes an amber warning. This helps users understand Bitcoin's position relative to this key technical level.
What happened to Alpha Vantage data?
The system now uses Coinbase exclusively for all price data. This provides more reliable access without API key dependencies and ensures consistent calculations across all factors using a single, authoritative price source with 700+ days of history.