Daily Alpha Brief — 2026-08-29 — BULL regime across majors

Bayesian crypto brief for 2026-08-29: BULL regime, strategy performance, weight shifts, and quant-vs-crowd sentiment across 9 symbols.

TL;DR

• Dominant regime: BULL (5 bullish / 4 bearish of 9 symbols).
• Top signal: volume_spike on BNB cut 0.310 (0.59 → 0.28).
• Shadow portfolio: $738 (-26.19%), 1 open, 35% win rate.

Shadow portfolio value since inception (dashed line = $1,000 starting capital).

Current Bayesian reliability weights, averaged across the universe.

Detected regime per symbol (EMA200 + ADX); % = lead-strategy conviction.

🌐 Market Indices

Crowd reads extreme greed at 81/100; BTC dominance 59.6% signals balanced.

🌐 Market Overview

Economic (FRED):

Stablecoin dominance 9.5% (risk-on); yield curve regime Flat.

Funding Rates (contrarian — heavy positive = crowded longs to fade):

Economic Calendar: Next FOMC: in 18 days | Next CPI: in 13 days | position-size modifier 1.00×.

📊 Regime Analysis

Regime per symbol from EMA200 + ADX on daily candles, with the fusion signal.

The book is broadly bullish: 5 BULL / 4 BEAR of 9 symbols.

🔗 Correlation Signals

Stable pairs (r > 0.70): SUI/AVAX: r=0.91 ✅ | BNB/SUI: r=0.87 ✅ | DOGE/SUI: r=0.86 ✅ | BNB/AVAX: r=0.86 ✅ | BTC/ETH: r=0.86 ✅ | XRP/SUI: r=0.85 ✅

Index signals: BTC dominance NEUTRAL → no rotation; Fear & Greed contrarian NONE.

⚙️ Strategy Performance

Aggregate 7-day performance per strategy (replayed across the universe). Edge = win-rate × avg P&L per trade — the table is sorted by it.

All P&L figures normalized to a 100 USDT position size for comparability (not summed across symbols).

🧠 Bayesian Weight Shifts

BNB/USDT (BULL):

In a confirmed BULL regime, ema_cross earns the highest trust: 43% win / 3.85 Sharpe / +121 pnl on 7d with strong 24h confirmation, nudged up modestly from its already-elevated 0.855 but capped given it doesn’t clear the 55% win threshold; walk_forward_ema similarly shows positive Sharpe (2.49) and pnl (+75) and gets a mild boost. rsi_macd has a troubling 24h (22% win, deeply negative Sharpe) that conflicts with its positive 7d Sharpe, so I cut it from 0.775 to 0.62 — the 24h collapse suggests deteriorating conditions even if the 7d aggregate looked acceptable. Bollinger is a clear rule-2/3 casualty (24% win, -12.21 Sharpe, -544 pnl on 7d) cut to near-floor; volume_spike (30% win, -5.70 Sharpe, -254 pnl) and tsmom (18% win, -12.82 Sharpe, -365 pnl — both windows deeply negative) are both cut hard per rules 2-3; post_shock_reversal’s 7d Sharpe of 6.64 and +373 pnl are impressive but it is a mean-reversion strategy conflicting with the BULL regime and its 24h is ugly, so I hold it near neutral; sentiment_strategy is nudged slightly down given zero 24h source coverage and near-zero composite signal; correlation_strategy and macro_strategy report no activity and are held near their current weights per the no-data guidance.

DOGE/USDT (BEAR):

volume_spike is catastrophically bad with 7d win_rate of 16% and Sharpe of -14.27 on 63 trades — easily the clearest rule-2/3 cut of the night, slashed to 0.10. walk_forward_ema also fires rules 2-3 (14% win, -5.56 Sharpe, -201 pnl on 7d with 7 trades just clearing the low-sample guard), cut to 0.10. post_shock_reversal has positive 7d pnl and a high Sharpe but its 7d win_rate is only 28% (rule 3 conflict: large wins masking frequent small losses) in a BEAR regime where mean-reversion faces trend headwinds, so it earns only a small nudge to 0.27 rather than a full boost. rsi_macd, ema_cross, tsmom, bollinger, correlation_strategy, and macro_strategy all have zero trades across both windows and are held within ±0.05 of current weight per rule 1; sentiment_strategy had near-zero composite and sparse 24h coverage so it is trimmed modestly from 0.639 to 0.60 rather than moved aggressively.

XRP/USDT (BULL):

ema_cross and walk_forward_ema are deeply penalised under rules 2-3: both show 7d win-rate of 25% and a catastrophic Sharpe of -8.27 with -616 pnl, contradicting the BULL regime prior (conflict rule 6) — cut to 0.15. volume_spike is even worse (22% win, -12.94 Sharpe, -1228 pnl) and cut to near-floor 0.10. post_shock_reversal similarly scores 28% win with -7.26 Sharpe and -840 pnl, so it is reduced to 0.15 despite its mean-reversion nature being regime-neutral. bollinger has positive 7d Sharpe (3.01) and pnl (+275) but a 38% win-rate falls below the rule-4 threshold, so it is trimmed modestly from 0.588 toward 0.45 — still the highest price-based weight given its positive risk-adjusted return. rsi_macd has only 3 7d trades (rule 1 low-sample guard) so it holds near its current 0.249. tsmom has zero trades (rule 1) and stays near current 0.336. sentiment_strategy has 3 sources with 100% confluence at a mildly positive composite, so it is held near its current elevated weight at 0.65 rather than penalised; correlation_strategy and macro_strategy are both zero-trade non-price voters held at their current weights of 0.263 rounded to 0.26.

📡 Signal Outlook

💼 Shadow Portfolio

Open positions:

$1,000 USDT paper portfolio, 1% risk/trade, MEXC fees modelled. No real capital is deployed.

📡 Discourse Radar

Sentiment analysis pipeline launching soon.

Academic Footnote

Fang et al. (2020) — a survey of cryptocurrency trading strategies.

Disclaimer

This is algorithmic analysis only, not financial advice.