Single-instrument long-short price-action exhaustion strategy on BTCUSDT.BINANCE that reads MULTI-BAR sequential persistence rather than single-bar anatomy. When BTC's 4-hour bars produce 5 or more consecutive same-direction bars (5 greens in a row = 20 hours of pure upward movement, or 5 reds in a row = 20 hours of pure downward movement) AND the cumulative percent move across the streak exceeds 2.5%, the directional flow is statistically exhausted: marginal momentum traders have already committed, contrarian capital is accumulating at the extreme, and a mean-reversion bar typically prints within the next 4-12 hours. SHORT the next bar's open after a 5+ green-bar streak with cumulative +2.5% or more; LONG the next bar's open after a 5+ red-bar streak with cumulative -2.5% or more. ZERO SUPPLEMENTARY DATA, ZERO EXTRA INSTRUMENTS, ZERO EXTRA BAR TYPES — purely OHLCV on the BTCUSDT.BINANCE-4-HOUR path (the same data path used by the landed BtcFourHourVolumeBreakoutLong, the most proven path in the catalog). Filling three under-represented portfolio buckets: (1) long-short direction (currently 9.8% vs ≤55% long-only ceiling — most-violated quota); (2) multi-bar pattern recognition mechanism class (every prior in-pipeline price-action hypothesis I submitted — BtcRangeExpansionExtremeCloseRejection — reads SINGLE-bar anatomy; this hypothesis reads cross-bar sequence persistence, fundamentally different signal class even though both are OHLCV-only); (3) sequence-based exhaustion vs the existing portfolio's volume-breakout, golden-cross, EMA-trend, multi-week-continuation, drawdown-accumulation strategies. Expected ~30-60 entry cycles per year × 6 years of BTC 4H data ≈ 180-360 trades, well above walk-forward sample-size floor. Mechanism is BTC-specific (calibrated to BTC 4H typical streak distribution), structurally immune to AVAX-style cross-asset analog overfitting since no claim of family-portability is made.