Deep dives into our methodology, the indicators we track, and the machine learning analysis behind our signal.
The Q1 2026 13F filings just dropped. Jane Street cut its IBIT position by approximately 71% and FBTC by approximately 60%. The basis trade that absorbed eighteen months of ETF inflows is being decommissioned by its largest operator. Future demand has nowhere quiet to land. Price discovery is back on the menu.
Read full article →The in-kind ETF model hands Authorized Participants control over Bitcoin sourcing. Because a liquid CME futures market exists, APs can hedge with paper contracts and delay spot purchases indefinitely. This structural plumbing problem is temporary. OTC supply is draining. Mining cannot keep pace. When sourcing channels run dry, suppressed demand hits the open market all at once.
Read full article →How the composite signal is built, tested, and validated.
5,790 daily observations, 5 verdict states, one transition matrix. The score has lived 86% of its days in STRONG BUY or ACCUMULATE. Half of all transitions are reversed within three days. When real changes happen, profit-taking ends in a V-shaped jump back to ACCUMULATE 70% of the time.
Read full article →We computed the correlation matrix across all 13 dashboard indicators. Within clusters, many are highly correlated. We added independent indicators, tested cluster correction, and ultimately let SHAP's original rankings stand after backtesting proved them right.
Read full article →Bitcoin's famous price-vs-time chart is a power-function regression, not a statistical Power Law. But the daily return tail (index ~2.7) and volatility autocorrelation (decays as lag^-0.33) are genuine Power Laws. The label is on the wrong object.
Read full article →13 indicators, each graded on a 0-100 scale, combined with SHAP-derived ML weights into a single score. From raw data to grade points to weighted average to verdict.
Read full article →XGBoost and LSTM models trained on 4,323 daily observations across 4 market cycles. Walk-forward validation, honest results, and the feature importance rankings that drive the dashboard weights.
Read full article →Does the signal actually work? Historical evidence.
Three DCA strategies compared across 12 entry points and 8+ years of data. Standard DCA vs Power Law vs Signal DCA. What works, what doesn't, and why.
Read full analysis →Pick any start date and base amount. Compare Standard DCA, Power Law DCA, and Signal DCA side by side with real historical data.
Run the backtest →6 simulated scenarios from bull run to 75% crash. 3 DCA strategies through each. The signal amplifies exposure — that's a feature and a risk.
See the scenarios →Academic papers and institutional research that informed our approach.
LSTM + SHAP analysis by the Bank of Spain. Key finding: feature importance shifts across market cycles. Technology variables dominated early; attention variables dominate now.
Read paper (PDF) →Comprehensive review finding that most models "perform only marginally better than random guesses" — validating our honest expectations about prediction difficulty.
S. Khedr et al., Decision Analytics Journal (2025) — Open Access →