Daily Alpha Brief — 2026-06-17 — BEAR regime across majors

Bayesian crypto brief for 2026-06-17: BEAR regime, strategy performance, weight shifts, and quant-vs-crowd sentiment across 10 symbols.

TL;DR

• Dominant regime: BEAR (0 bullish / 8 bearish of 10 symbols).
• Top signal: walk_forward_ema on PEPE cut 0.283 (0.50 → 0.22).
• Shadow portfolio: $978 (-2.18%), 5 open, 43% 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.

📊 Regime Analysis

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

The book is broadly bearish: 0 BULL / 8 BEAR of 10 symbols.

⚙️ 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

PEPE/USDT (BEAR):

Bollinger is the standout performer with 9 trades at 89% win-rate and a Sharpe of 16.55 over 7d, earning a strong raise per rule 4; in a BEAR regime mean-reversion after band pierces also gets regime support, so I raised it to 0.72. walk_forward_ema is the clearest loser: 7d win-rate of 19% (rule 2 fires), deeply negative Sharpe of -12.79 and pnl of -818, and as a momentum strategy it conflicts with the BEAR/ranging data (rule 6), so I cut it aggressively to 0.10. volume_spike shows a severe 7d contradiction — 30% win-rate with a Sharpe of -4.38 and -373 pnl despite a good 24h snapshot, triggering rules 2-3 and a cut to 0.20. post_shock_reversal has strong 7d metrics (Sharpe 9.95, +646 pnl) and suits the BEAR/post-shock context so I raised it to 0.62; rsi_macd and ema_cross have zero trades (rule 1 thin-data guard, held at 0.50); tsmom has exactly 5 7d trades so the guard keeps it near 0.50 despite modest positive metrics; sentiment is held near its current weight at 0.38 given weak 7d confluence of only 33% and near-zero composite signal.

BNB/USDT (SIDEWAYS):

Bollinger is the standout performer in this SIDEWAYS regime with 7d win_rate 64%, Sharpe 13.42, and +304.67 pnl, earning a strong raise per rule 4 and regime alignment; the extraordinarily high Sharpe suggests low volatility of returns which supports further trust. post_shock_reversal has strong 7d pnl (+267.39) and Sharpe (7.87) but a below-50% win_rate (39%), indicating a few large wins carry the result — it is a mean-reversion family strategy that suits SIDEWAYS, so a modest boost is warranted while the sub-50% win_rate caps the move. volume_spike shows conflicting signals: a brutal 24h Sharpe (-6.74) with a modest positive 7d Sharpe (1.44) and very low win_rate (35%), which is below the 30% cut threshold but not both windows simultaneously negative on pnl, so a measured reduction applies with the SIDEWAYS small bump partially offsetting. rsi_macd, ema_cross, tsmom, and walk_forward_ema all have zero or near-zero 7d trades (walk_forward_ema has only 2), triggering rule 1 for all — held near current weight; sentiment shows weak 7d confluence (67%) on a mildly negative composite (-0.09) with only 3 sources, warranting a small reduction from its already-below-neutral 0.40.

BTC/USDT (BEAR):

post_shock_reversal is the standout performer with 52 trades, 48% win rate, and an extraordinary 7d Sharpe of 15.93 with +575 pnl, earning a significant boost per rule 4; in a BEAR regime post-shock mean-reversion is well-supported empirically and the data confirms it. volume_spike also shows strong, high-sample 7d metrics (79 trades, 8.45 Sharpe, +381 pnl) and earns a meaningful raise despite below-50% win rate due to its regime-agnostic nature and exceptional risk-adjusted return. tsmom fires rule 3 with a negative 7d Sharpe (-0.89) and sub-30% win rate on 22 trades, conflicting with the BEAR regime’s momentum lean, so data wins and it is cut. rsi_macd and ema_cross have zero trades in both windows (rule 1: hold within ±0.05 of current); walk_forward_ema has only 3 7d trades so it is held near neutral per rule 1; bollinger has borderline low win rate (33%) despite a positive Sharpe in BEAR context so it is nudged slightly down; sentiment shows weak confluence (33-50%) and mildly positive but near-neutral composite, so it is reduced modestly from its already-below-neutral starting weight.

📡 Signal Outlook

💼 Shadow Portfolio

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

📡 Discourse Radar

Crowd-vs-quant sentiment from CryptoCompare news, crypto RSS, the macro feeds (FT / Fed / CNBC), and the Fear & Greed index. (Reddit retired — Responsible Builder Policy restrictions, 2026.)

• Fear & Greed: -1.00 (bearish crowd).

Most-talked alt: DOGE/USDT at +0.94 (bullish crowd lean).

📰 Today’s Key Reads

Curated from 3 sources monitored today:

🟢 Trump Tariffs 3: Return of the Bull Market! NYSE Tokenising, what that means for $Hype! Claude Meme Meta! — decrypt — Score: +0.94

🟢 Tom Lee’s BitMine Adds $139 Million in Ethereum Ahead of Preferred Share Trading Start — decrypt — Score: +0.81

🟢 Trump’s speech at Davos gives small boost to Crypto! Saylor buys $2.13B BTC! Blondish Interview! — decrypt — Score: +0.65

🟢 Moody’s rolls out credit ratings on Solana in tokenized asset push — coindesk — Score: +0.62

🟢 Bitcoin doesn’t need Ethereum-style yield, says Strategy’s Michael Saylor — cointelegraph — Score: +0.57

Academic Footnote

Krauss (2017) — statistical arbitrage pairs trading.

Disclaimer

This is algorithmic analysis only, not financial advice.