Daily Alpha Brief — 2026-08-28 — BULL regime across majors

Bayesian crypto brief for 2026-08-28: BULL regime, strategy performance, weight shifts, and quant-vs-crowd sentiment across 9 symbols.

TL;DR

• Dominant regime: BULL (6 bullish / 3 bearish of 9 symbols).
• Top signal: ema_cross on WIF raised 0.377 (0.22 → 0.59).
• Shadow portfolio: $739 (-26.15%), 0 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 82/100; BTC dominance 59.7% signals balanced.

🌐 Market Overview

Economic (FRED):

Stablecoin dominance 9.4% (risk-on); yield curve regime Flat.

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

Economic Calendar: Next FOMC: in 19 days | Next CPI: in 14 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: 6 BULL / 3 BEAR of 9 symbols.

🔗 Correlation Signals

Stable pairs (r > 0.70): SUI/AVAX: r=0.89 ✅ | BNB/SUI: r=0.87 ✅ | DOGE/SUI: r=0.85 ✅ | DOGE/AVAX: r=0.85 ✅ | BTC/ETH: r=0.84 ✅ | SOL/SUI: r=0.83 ✅

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

WIF/USDT (BULL):

rsi_macd and tsmom have zero trades on both windows (rule 1 low-sample guard), so they stay within ±0.05 of current; the BULL regime nudges them marginally up from 0.217. bollinger has only 3 trades on 7d (rule 1) and a 33% win-rate in a BULL regime where band-walk risk is elevated, so a small downward nudge from 0.334 to 0.30 is warranted. ema_cross similarly has just 1 trade (rule 1 applies), so despite the perfect win-rate the data is noise and we hold close to current with only a tiny BULL nudge. volume_spike has strong 7d sample size (72 trades) but a 33% win-rate (borderline rule 2 territory) and a Sharpe that is inflated by pnl asymmetry; the positive pnl saves it from a hard cut but the win-rate drags it down from 0.401 to 0.38. walk_forward_ema has 28 7d trades with a 39% win-rate and positive pnl/Sharpe — not strong enough for rule 4 but no rule 2/3 trigger either, so it holds near current with a modest BULL regime endorsement at 0.43. sentiment_strategy shows near-zero composite and 0% 24h confluence with sparse 2-source coverage, warranting a small downward nudge from 0.490 to 0.47. correlation_strategy and macro_strategy both report zero activity and hold at their current weights per rule 1. post_shock_reversal has 10 7d trades with a 30% win-rate and a positive but weak Sharpe; it conflicts with the BULL regime (mean-reversion family), so it is nudged slightly down to 0.17.

BNB/USDT (BULL):

In a confirmed BULL regime, ema_cross holds near its current high weight (7d 41% win but positive Sharpe 2.35 and +79 pnl, regime-aligned momentum) while rsi_macd is trimmed from 0.822 toward 0.72 because its 7d win-rate of 43% is below the 55% threshold for a full boost despite its high Sharpe, and its 24h win of 33% is weak. Bollinger is cut further to 0.10 per rule 3 (7d Sharpe -2.18, pnl -97.14) — classic band-walk casualty in a BULL trend. Tsmom had only 5 trades on 7d (rule 1 thin-data guard) but both windows show negative Sharpe and negative pnl, so a modest cut to 0.55 is warranted without a large move; post_shock_reversal is a mean-reversion strategy struggling in a trending market (7d 39% win) so it is nudged down slightly; walk_forward_ema earns a small cut from 0.817 to 0.72 because its 24h Sharpe is deeply negative (-4.33) even though 7d metrics are modestly positive, signalling recent deterioration. Sentiment gets a marginal bump on moderate 7d confluence (67%), while correlation and macro hold near their current weights as they show no active signals.

SOL/USDT (BULL):

In a confirmed BULL regime, ema_cross and walk_forward_ema are penalized by rule 2/3: both show 7d win-rate of 32% with negative Sharpe (-1.82) and negative pnl (-129.40), a clear data-over-regime conflict (rule 6), so both are cut to ~0.14. tsmom is severely punished — 7d win-rate of only 20% with Sharpe -5.38 and pnl -351.05 triggers rule 2 (persistent underperformance), dropping it to 0.20 despite the BULL prior. bollinger earns a modest raise to 0.65 on strong 7d metrics (48% win, Sharpe 14.31, +1044 pnl), and volume_spike stays elevated at 0.60 given positive 7d Sharpe and pnl despite a weak 24h; sentiment_strategy is trimmed slightly from 0.599 to 0.58 because the avg_composite is near-zero (+0.02), indicating minimal directional conviction despite strong confluence. correlation_strategy and macro_strategy both report zero trades across all windows and are held near their current weights per the no-data-hold principle.

📡 Signal Outlook

💼 Shadow Portfolio

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

📡 Discourse Radar

Sentiment analysis pipeline launching soon.

📰 Today’s Key Reads

Curated from 3 sources monitored today:

🟢 Galaxy expands retail crypto lending with new BTC, ETH and SOL-backed credit line — theblock — Score: +0.68

🟢 SOL treasury firm DeFi Development launches real-time ‘State of Solana’ intelligence platform — theblock — Score: +0.67

⚠️ Bitcoin treasury companies shed $80bn in value as business model unwinds — ft_markets — Score: +0.49

🟢 Bitcoin holds $79,000, ether, solana slip as traders bank a week of gains — coindesk — Score: +0.34

🔴 Bitcoin below $79,000, XRP leads losses as traders start betting on a Fed hike — coindesk — Score: -0.34

Academic Footnote

Tzouvanas et al. (2020) — moving-average trading rules for Bitcoin.

Disclaimer

This is algorithmic analysis only, not financial advice.