When Should You Buy Bitcoin
A quantitative approach to identify Bitcoin buying opportunities using two independent valuation metrics.
🎯 The Strategy
Bitcoin enters a Double Undervaluation Buy Zone when BOTH conditions are met:
- Price < 200-Day DCA Cost - Current price is cheaper than the average cost of consistent buyers over the past 6-7 months
- Price < Power Law Trend - Current price is below Bitcoin's long-term fundamental value based on network growth effects
Using two independent metrics reduces false signals and identifies stronger buying opportunities.
📊 Website
When Should U Buy Bitcoin
Features
- Historical Charts - Visualize when Bitcoin has been in the buy zone
- Real-Time Check - Get current valuation status
- Trend Outlook - Power-law trend value 1, 2 and 4 years ahead, and how much it moved since the fit of a year ago (Charts view)
- Strategy Backtesting - Test different investment strategies with historical data
- Distance to Buy Zone - See how much Bitcoin needs to drop to enter the zone
Data Updates
- The daily update is run by the DCA service scheduler, which runs
main.py every day at 00:30 UTC to refresh docs/data/ and the charts
- The same regeneration also runs after a successful DCA transaction and from the admin-only regenerate button on the stats page; only one run is active at a time, and a run exceeding 30 minutes is terminated
- Output of each run is appended to
data/static_generation.log
- Scheduler settings (environment or
.env): STATIC_GENERATION_SCHEDULE_ENABLED (default true), STATIC_GENERATION_SCHEDULE_HOUR_UTC (default 0), STATIC_GENERATION_SCHEDULE_MINUTE_UTC (default 30); set STATIC_GENERATION_STRICT_UPDATE=true to fail the run if any chart or report update step fails
- The GitHub Actions workflow
update-data.yml is manual trigger only (workflow_dispatch); its cron schedule is intentionally commented out
- Real-time check fetches live BTC price from Binance first, then falls back to Coinbase
- Historical charts and daily summaries use the generated files in
docs/data/; check each page's displayed update timestamp before treating non-price metrics as current.
Local DCA Service (Poetry)
- Start/restart backend on port 8000:
lsof -i :8000 | awk 'NR!=1 {print $2}' | xargs kill -9; poetry run uvicorn dca_service.main:app --reload
- Run DCA service tests with Poetry:
poetry run pytest dca_service/tests -v
Macro Risk Score (Quick Config)
The chart Macro Risk Score is a weighted blend of macro stress signals:
- Net Liquidity 90D change
- SOFR
- MOVE
- HY OAS
Default weights are defined in:
src/whenshouldubuybitcoin/visualization.py -> MACRO_RISK_SCORE_WEIGHTS
Current defaults:
net_liquidity_90d_change: 0.35sofr: 0.20move: 0.20hy_oas: 0.25
To quickly tune weights:
- Edit
MACRO_RISK_SCORE_WEIGHTS in visualization.py
- Regenerate charts:
poetry run python main.py
🎮 Strategy Backtesting
Test different Bitcoin investment strategies without risking real money. All calculations run entirely in your browser.
Available Strategies
1. Daily DCA
Invest the same amount every day. Simple and consistent.
2. Monthly DCA
Invest your full monthly budget on a specific day each month (1-28). More realistic for most investors.
3. AHR999 Historical Percentile
Dynamically adjust investment based on where current AHR999 ranks historically. Buy more when Bitcoin is cheaper.
Default Multipliers:
- Bottom 10% (EXTREME CHEAP): 5x daily investment
- 10-25% (Very Cheap): 2x daily investment
- 25-50% (Cheap): 1x daily investment
- 50-75% (Fair): 0x (no investment)
- 75-90% (Expensive): 0x (no investment)
- Top 10% (VERY EXPENSIVE): 0x (no investment)
Features:
- Fully customizable multipliers for each percentile tier
- Real-time display of daily investment amounts
- Unlimited budget mode option (ignore total budget constraints)
- Only available for historical backtests (requires AHR999 data)
Results
- Total invested, final BTC balance, portfolio value
- Total return (%) and annualized return (%)
- Interactive chart showing portfolio growth over time
- Detailed transaction history with AHR999 values
🔬 Methodology
200-Day DCA Cost
DCA Cost = 200 / Σ(1/price_i)
Harmonic mean of the last 200 daily closing prices.
Power Law Trend Model
Trend = a × days^n
Fitted via linear regression on log-log transformed data. Models network effects (Metcalfe's Law) with decreasing growth rate over time.
⚠️ Disclaimer
This tool is for educational purposes only. Do your own research before making investment decisions.
