Hypotheses

Derived by the Research Lead from Martin Zweig's breadth-thrust indicator. Zweig's idea: when market breadth moves from oversold to broadly advancing within about 10 sessions, a lasting advance is starting. Here it is applied to a basket of crypto alts. The knowledge base has no test of a breadth-THRUST event. The only related test used continuous breadth-scaled allocation, which is a different rule. Universe: 12 alt spot markets on OKX_SPOT with daily bars from 2020: SOL, XRP, ADA, DOGE, LINK, LTC, DOT, AVAX, BCH, ATOM, ETC, FIL. Selection rule: the largest non-BTC, non-ETH, non-stablecoin alts with an OKX USDT spot market listed by 2020-12. The universe was fixed before any test. Point-in-time handling: on each daily close a name is ELIGIBLE only if it is listed (is_listed / universe_at) and its SMA has a full window. Breadth and entries use eligible names only. Breadth is computed only when at least min_eligible=8 names are eligible. Remaining survivorship bias: OKX keeps no delisted spot pairs, so a name that later died cannot be included. The report must say so. Signal: breadth_t = fraction of eligible names whose daily close is above their own sma_len-day SMA. A THRUST fires on daily close t when all three hold: breadth_t >= thrust_high; breadth_{t-1} < thrust_high; and min(breadth over the prior thrust_window closes, excluding t) <= thrust_low. On a thrust, while flat, buy every eligible name in equal notional. Execution: decisions are made on the 1-DAY close. Orders are market orders at the signal bar's close. On a 24/7 venue that price is the next bar's open (registry next_bar_open_fill: use_equivalent). All legs exit together after hold_days daily closes, or earlier on the first daily close where breadth < exit_breadth. No new thrust is taken while the basket is held. Risk control route (b): signal/time exit with capped sizing and no stop-loss. Each leg = gross_exposure / N_eligible of equity (60% / 12 = 5% per leg). Exposure never exceeds 100% of equity; unlevered cash account. Pre-study on local OKX_SPOT daily bars 2020-2026. It is gross, close-to-close and before costs. It was NOT a decision input and was used only to set predictions. - With sma 50, low 0.25, high 0.75, window 10, the thrust fires about 29 times in 6.5 years (about 4-5 per year), in every calendar year including 2026. - Mean 20-day equal-weight basket return after a thrust is +9.6% (median +5.2%). Unconditional: +4.4% (median +0.0%). - With the breadth<0.35 exit, the 30-day cap and 60% gross: 354 leg-trades, mean leg return about +12% but median about -3.5%. The edge is right-tailed: 2021 and 2024 dominate. Max drawdown about 25%; 2022, 2025 and 2026 were negative. - The SMA-20 variant was much worse (drawdown 54%), so sma_len is the key sensitivity. Fee arithmetic: an OKX spot round trip costs about 0.20% of notional. The mean leg captures several percent, carried by trend legs, while the median leg loses. Honest predicted average is about 3% per leg, more than 10x the cost.

Cross-venue, market-neutral basis trade between a redeemable gold token and the metal itself. PAXG (Paxos Gold) is a fully backed ERC-20 token: 1 PAXG = 1 fine troy oz of London Good Delivery gold, which Paxos mints and redeems. PAXGUSDT on Binance spot is priced by crypto order flow. XAUUSD.SPOTMETAL is spot gold (Dukascopy MID). When crypto holders dump PAXG for liquidity (risk-off deleveraging, stablecoin demand, exchange-specific outflows), the token can trade at a transient discount to the gold it represents. Mint/redeem arbitrage and gold-motivated crypto buyers pull it back. The strategy buys PAXG spot (long-only CASH leg, the only side spot allows) and shorts an equal notional of XAUUSD.SPOTMETAL (margin leg), so gold-price exposure cancels and only the basis is held. It exits when the discount closes. The basis is measured only when spot gold is in session: a fresh XAU 1H bar must share the PAXG bar's close timestamp. Weekend and 17:00-18:00 NY break hours are skipped for signals, because PAXG trades then and XAU does not, and that apparent basis is stale-price noise. Execution: PAXG is a 24/7 venue, so its market order fills at the signal bar's close (the faithful equivalent of the next bar's open). XAUUSD is a session venue, so its order fills at the next 1H bar's open (next_open_fill, supported). The legs can be up to one hour apart; this is acknowledged as hedge slippage. Neither PAXG nor SPOTMETAL has ever been used in this factory (SPOTMETAL is 0.3% of experiments, PAXG 0%). The trade fills the under-represented cross-venue (9.3% vs the >=15% target) and long-short (23% vs 45%+) buckets and is no lead-lag claim: the two legs are claims on the same ounce. The test is limited to the discount side. Shorting PAXG at a premium would need the PAXGUSDT perp, which only has bars since 2025-03, so it is excluded to keep the full 2020-08+ history.

This is the upside mirror of the factory's promoted OKX hourly capitulation reversal. It trades the opposite forced flow, on a different venue (Bybit linear USDT perpetuals). It enters on an event and exits on a fixed clock. A SHOCK is a single 1H bar where both of these hold: (1) the log return is at least shock_sigma standard deviations of the coin's own last 720 hourly returns (30 days); (2) the log volume is at least volume_sigma standard deviations above its own 720-bar mean. We treat that move as a short squeeze, forced short covering or a FOMO taker sweep. That buying comes from liquidation engines and stop orders, not from information. It overshoots the price at which voluntary liquidity providers will hold the coin. The strategy supplies liquidity on the other side. It waits one full hour, skipping the squeeze's immediate continuation bar. It then sells the perp short at the confirmation bar's close and buys it back on a fixed 24-hour clock. It is flat about 95% of the time. The universe was fixed in advance, not chosen by performance: the 12 Bybit USDT perps that were already liquid top-30 alts with Bybit linear listings by mid-2021 (SOL, AVAX, LINK, DOT, ADA, DOGE, XRP, LTC, ATOM, FIL, UNI, NEAR). Point-in-time handling: parameters.point_in_time_universe=true, so each leg trades only from its own first Bybit bar. Remaining survivorship bias: Bybit serves no bars for closed contracts, so all 12 names are still listed. Execution: this is a 24/7 venue, so 'next-bar open' is written as the platform equivalent, a market order at the confirmation bar's close (next_bar_open_fill = use_equivalent). The protective stop is an intrabar reduce-only stop_market. The platform's simulated venue triggers it inside the bar in backtest and paper, and live trading mirrors those fills. No exchange-resting order is needed. Sizing: notional = min(notional_fraction x equity, risk_per_trade x equity / stop_distance_pct), with at most max_concurrent open events across the panel. Fee arithmetic: a >=4-sigma hour on these alts is a ~3.5-6% rise. If 20-30% of it retraces over 24h, the gross capture is ~0.8-1.4%. Costs are Bybit's 0.11% round trip plus ~0.05-0.10% impact, so the expected net is ~0.35-0.5% per trade, more than 2x the cost. Squeeze funding is usually positive, so the short typically receives it; this is not part of the edge claim. Why this is new: the OKX sibling and the Bybit volume-share survivor are LONG trades after down-shocks. Prior upside-shock tests were long continuation ('Delayed Second-Leg After an Alt Upside Shock' died premise_flawed).

Derived by the Research Lead from Dmitriy Gizlyk’s [Practical use of clustering](https://www.mql5.com/en/articles/10943): form price/activity clusters without return labels, then estimate subsequent returns on a separate chronological sample. The source supplies this modeling approach, not these trading rules or evidence of profitability. Trade ALGO and ICP independently on 4-HOUR bars, long, short or flat. Market orders at signal-bar closes implement the registry’s faithful equivalent of next-bar-open execution on continuous markets. Use all available history separately for each instrument after causal warm-up; do not assert equal listing dates or a full 2.3-year span. The fixed algorithm refreshes learned states every 30 UTC days; frozen operation means its rules and hyperparameters remain fixed, while this prescribed causal state update continues. No pre-study or performance-based screening was run.

Derived by the Research Lead from Da, Gurun and Warachka's Frog in the Pan hypothesis: test whether gradual advances subsequently outperform equally large jump-driven advances. Source: [Frog in the Pan: Continuous Information and Momentum](https://academic.oup.com/rfs/article-abstract/27/7/2171/1578455). Only the published abstract was consulted; its equity findings do not establish a crypto edge. Trade one equal-dollar long-short pair selected from six named Hyperliquid assets using 4-HOUR bars and internally aggregated UTC daily returns. Execute at completed signal-bar closes, the supported equivalent of next-bar-open execution on continuous venues. Risk-control route (b): fixed-clock exits, 15% MTM-equity notional per leg, 30% gross exposure and leverage 1. Request all available history, acknowledging Hyperliquid's approximately 5000-candle limit. No preliminary performance test was run.

Test short MAVIAUSD.HYPERLIQUID against equal-base-quantity long MAVIAUSDT.BYBIT using synchronized 4-HOUR bars. The funding snapshot contains 980 overlapping daily observations from 2024-02-07 through 2026-10-13, with mean Hyperliquid-minus-Bybit settled funding of 0.0679019% daily. This is descriptive evidence, not a backtest. Instrument selection after comparing funding histories must count toward multiple-testing accounting. Market orders at synchronized signal-bar closes implement the registry's equivalent of next-bar-open execution on continuous venues. Normalize actual settlement payments by elapsed time: Hyperliquid settles hourly ([documentation](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/funding)); Bybit intervals are instrument-specific ([documentation](https://bybit-exchange.github.io/docs/v5/market/history-fund-rate)). Use the existing engine funding ledger without adding duplicate credits.

Test whether ETH option insurance purchased before scheduled FOMC announcements is overpriced relative to realized announcement risk. Sell a protected, same-expiry Deribit iron condor before the announcement and close afterward. Use actual option prices and completed 4-HOUR bars through RollingOptionsStrategy. Market decisions execute at the signal close, the supported continuous-market equivalent of next-bar-open execution. ETH-PERPETUAL.DERIBIT supplies the clock and spot reference and is not traded. Expand all four historical option IDs into extra_instruments before verification. Obtain scheduled announcement dates and release timestamps from the [Federal Reserve calendar and linked historical statements](https://www.federalreserve.gov/monetarypolicy/fomccalendars.htm). The calendar source does not supply this trading strategy.

Derived by the Research Lead from Tushar Chande's Trend Score as published on ProRealCode (Source URL: https://www.prorealcode.com/prorealtime-indicators/tuschar-chandes-trend-score/). The source defines the score: compare today's close with each close from k=11 to k=20 bars ago, score +1 if today is higher and -1 if lower, and sum the ten votes (range -10..+10). The source then smooths the score (SmoothingPeriod=5). The source is an indicator only; the trading rules below are the Research Lead's operationalization. Mechanism: the score is a sign-vote on the medium-horizon trend that deliberately skips the most recent 10 bars. It measures how persistent the trend is and ignores the short-term reversal noise that kills raw-momentum crossovers on alts. Market: five Binance COIN-M inverse alt perpetuals that the corpus has never tested (LINK, DOT, BCH, ETC and FIL _PERP.BINANCE_CM, 0 prior experiments each). BINANCE_CM is 2.6% of the corpus against a quota of at least 5%. Bars are 12-HOUR, and each perp is traded independently as its own netted position. Rules: go long when the smoothed score is >= +entry_threshold and short when it is <= -entry_threshold. Stay flat in between; the strategy is not forced to always hold a position. Exit when the score crosses back through the exit band, or on a platform-matched intrabar ATR trailing stop (a reduce-only _submit_stop_market; no exchange-resting order is required). Size each position with contracts_for_notional() on the inverse USD-face contracts at 18% of equity notional per perp, unlevered (max gross 90%). Fee arithmetic: on 12H alts, a trend that persists across the 11-20 bar (5.5-10 day) vote window usually spans a multi-day move of 6-15%. Capturing a third of that gives about 2.0-3.0% gross per trade. Costs are about 0.06% COIN-M round-trip taker plus about 0.05% impact and funding drag, so the expected net is >= 1.0% per trade, more than 10x cost. Expected frequency: about 10-15 trades a year per perp, 55-70 a year across the five, and 300+ over the 2020-09..2026-10 history. This is distinct from the many 4H/1D dual-timeframe EMA-confluence variants already in the corpus. It has no higher-timeframe filter, uses a lagged sign-vote instead of price versus a moving average, and trades on a coin-margined venue.

This tests a mechanism the factory has already promoted, but on a different exchange it has barely used. The promoted Binance strategy (SpotPerpVolumeShareCashAccumulationAltBasketLong: 213 trades, average +9.78% per trade, walk-forward OOS Sharpe 1.70) buys an alt on spot when spot's share of spot+perp volume is unusually high and the coin is in an uptrend. Its claim is that a rise paid for with cash lasts, while one paid for with leverage gets unwound. If that is really about who is buying, and not a quirk of Binance's venue mix, it should also show up in OKX's own order flow, which has a different user base and a different spot/swap ratio. Here the signal and the trades both use OKX data only: share = OKX spot base volume / (OKX spot + OKX USDT-swap base volume), traded long-only on OKX_SPOT with a 1x cash account. Fixed universe, chosen by a history rule before any returns were looked at: every non-BTC/ETH alt with both OKX spot and swap daily history starting on or before 2021-01-31. That gives SOL, XRP, ADA, DOGE, LINK, LTC, DOT, AVAX, ATOM, BCH, ETC and FIL; weak names stay in. PRE-STUDY (OKX public REST daily candles, 2020-01 to 2026-10, gross of costs, close-to-close): unconditional 5-day forward return +0.84% mean / -0.33% median. Trend-only control +1.70% / -0.27%. Joint rule (share z>1.6 and 77-day return >0) +3.62% / +1.21% over 1,205 coin-days, so the share filter roughly doubles the trend-only mean. Taking one position per coin at a time with a 5-day hold gives 525 events, mean +2.56%, median +0.55%, 52% win rate. By year: 2020 +7.5%, 2021 +4.8%, 2022 -1.5%, 2023 +2.9%, 2024 +3.6%, 2025 -1.7%, 2026 +2.0%. 10 of 12 coins were positive (ATOM and ETC slightly negative). Weakness, stated up front: it loses in alt bear years, and the holdout covers mostly 2025. Fees: about 0.25% round trip (0.20% taker plus impact) against about 2.5% gross per event. Execution: market orders at the daily signal close. On a 24/7 venue that price equals the next bar's open; this is the registry's use_equivalent for next_bar_open_fill. The developer must use O(1) rolling updates, because 24 daily series is about 55k bars. Expose require_share_filter (default true) so the matched trend-only control can be run with one flag.

Trade DASH and XLM Hyperliquid perpetuals independently on 4-HOUR bars, long and short. Test whether a breakout accepted through several consecutive closes subsequently produces a tradable unwind when price returns inside its original range. Market orders execute at the signal close, the registry's faithful equivalent of next-bar-open execution on continuous venues. Protective stops trigger intrabar through platform matching. Use all available exact-instrument history; DASH's shorter coverage limits confidence. No preliminary backtest was run.

Derived by the Research Lead from wufuhao100w's single-pair versus multi-level grid question (https://www.fmz.com/bbs-topic/4684): test its inventory-limitation implication using one outstanding entry and one non-pyramided position per instrument. The full local post was read; it supplies a question, not measured performance. Trade independent XLM and ETC 4-HOUR perpetual bars with wide passive entries, fixed protective stops and larger profit targets. Replace the source's opposing order pair with one eligible side to avoid assuming atomic cancellation. Passive limits and platform-matched intrabar stops are supported. Crypto market exits execute at the signal bar's close, the registry's equivalent to next-bar-open execution. Use all available history, approximately 2.3 years maximum per Hyperliquid 4h series.

Test whether downside volatility building beneath a nearly unchanged ETH price is underpriced by weekly downside insurance. Buy a same-expiry ATM put and sell an equal-quantity put approximately 6% below spot only when a downside-variance acceleration signal and a conservative premium ceiling agree. This is original autonomous research, not a source port. Extend RollingOptionsStrategy directly. ETH-PERPETUAL.DERIBIT is an untraded 4-HOUR clock; ETHUSD.HYPERLIQUID supplies the untraded 4-HOUR return history. Resolve actual historical Deribit ETH weekly puts through rolling_options, underlying ETH, weekly cadence, dte_target 7, with P offsets 0 and -0.06. The two listed reference instruments are not the option trading universe: append every resolved, full option instrument id to instruments/extra_instruments and the persisted execution contract before verification. Never invent a fixed expired option id or replace options with perpetual exposure. Use all genuinely overlapping history and one 4-HOUR stream per instrument. Orders execute at completed signal closes, the registry equivalent of next-bar-open execution on continuous venues. Actual option observations, not synthetic prices, determine entry and exit fills. Standard option fees are charged on underlying notional, not option premium; source: [Deribit Fees](https://support.deribit.com/hc/en-us/articles/25944746248989-Fees).

Derived by the Research Lead from [Convex Finance documentation](https://docs.convexfinance.com/convexfinance/general-information/understanding-cvx): test multi-day relative-price convergence between CRV and CVX. The source establishes governance and reward links, not a trading strategy or evidence of convergence. Trade synchronized 4-HOUR bars on CRVUSD.HYPERLIQUID and CVXUSDT.BYBIT. Market orders execute at the signal bar close, the registered faithful equivalent of next-bar-open execution on continuous venues. Use all common available history, reset indicators after gaps, and maintain only one paired position. This tests price convergence, not on-chain rewards or funding carry.

Derived by the Research Lead from the submitted industry-level high-anchoring hypothesis. Preserve daily bars, the 20-ETF candidate set, monthly ranking, equal-dollar longs in the three ETFs nearest their trailing high, and shorts in the three furthest below it. Repair the executable contract by requiring config.parameters.point_in_time_universe=true in initial, optimization, refit and replay configurations. Historical window selection must use the primary ETF's available history without truncating it to a later extra-leg inception. Each ETF joins only after its own lookback has warmed. Separate panel-dependent ranking from portfolio risk monitoring and outstanding-target management. Decisions at session close fill at the next session open, as supported by the supplied execution registry. These corrections preserve the intended experiment; advisory smoke-window losses do not establish failure over its intended history. The fixed surviving-ETF candidate list retains selection bias, and dividends and borrow costs are not fully represented.

Test a hedged funding-receipt regime on TRXUSD.HYPERLIQUID and TRXUSDT.BYBIT using synchronized 4-HOUR bars. Long the venue with persistently negative funding and short the venue with persistently positive funding, so both legs receive funding. Enter only when a haircut forecast of receipts covers commissions, impact and adverse basis convergence. Market orders execute at the signal-bar close, the registry equivalent of next-bar-open execution on continuous venues. Fills are sequential. Use full overlapping historical coverage, beginning no earlier than Hyperliquid funding coverage on 2023-08-06. Aggregate actual settled funding by elapsed time: [Hyperliquid funding](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/funding) settles hourly, while [Bybit funding-history documentation](https://bybit-exchange.github.io/docs/v5/market/history-fund-rate) requires instrument-specific intervals. Funding cash flows must come from the engine ledger without duplicate strategy credits.

Derived by the Research Lead from Binance's WBETH documentation: test whether staking rewards embedded in WBETH appreciation survive trading costs when ETH price exposure is hedged. Buy WBETHUSDT.BINANCE_SPOT and short ETHUSD.HYPERLIQUID using synchronized 1-DAY bars. USDCUSDT.BINANCE_SPOT is an untraded currency reference. No staking, redemption, transfers or atomic execution are assumed. Rewards enter through WBETH's traded price; never credit a separate staking cash flow. Market orders execute at the signal close, the registry's equivalent of next-bar-open execution on continuous markets. Historical conversion ratios are a declared data dependency. Source: [What Is WBETH?](https://www.binance.com/en-IN/support/faq/detail/e252366155174ba6887f6b32e3798273). Commission assumptions follow [Binance spot fees](https://www.binance.com/en/fee/trading) and [Hyperliquid fees](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/fees).

Derived by the Research Lead from ianzeng123's gold-token statistical-arbitrage framework: https://www.fmz.com/strategy/526812. Test PAXG against XAUT perpetuals using synchronized 4-HOUR bars. Retain the source's rich-token short/cheap-token long mechanism and frozen entry-mean exit; introduce slower sampling, cross-venue execution, a cost-aware deviation floor, capital-relative sizing and a holding limit. The linked video was not viewed. Signal-close market orders are the platform's equivalent of next-bar-open execution on these continuous venues. No independent opposing positions on one instrument, exchange-resting stops, queue priority or concurrent executor deadlines are required. Use all actual overlapping history, subject to Hyperliquid's approximately 5000-candle limit. Bybit contracts are documented in [Bybit's gold derivatives guide](https://www.bybit.com/en/learn/bybit-guide/bybit-gold-derivatives-trading); PAXG backing is described in [Paxos documentation](https://docs.paxos.com/guides/stablecoin/paxg).

Derived by the Research Lead from the supplied Lemperiere operationalization and analyst revision request: preserve the correctly implemented AVAXUSDT.OKX daily long-short signal, but halve daily_vol_target to 0.0075 and max_fraction to 0.5. This tests a revised allocation premise; it does not establish a new directional edge or guarantee proportional drawdown reduction. Risk-control route (b): signal exits, entry notional capped at 50% of equity, leverage 1.0, quantity fixed until exit. Continuous-market orders execute at the signal bar close, the supported equivalent of next-bar-open execution. Cost arithmetic is a hypothesis, not a measured result: assumed gross capture 0.50% minus 2 × 0.05% configured OKX taker commissions leaves 0.40% before slippage and funding; predict 0.30% net per trade. Assumed gross capture is five times commission cost. Preserve all previously exposed development, OOS and inspected-holdout evidence in the existing lineage.

Derived by the Research Lead from 发明者量化-小小梦, 基于动态时间规整的形态识别策略: test daily DTW matching against an explicitly specified recovery template on Chinese A-shares. Preserve the source's long-only selection, five-session holding period and exit after a daily fall exceeding 5%. The source omits its actual template, normalization, distance threshold and allocation; this pilot supplies those choices without attributing them to the author. Use 1-DAY bars, a baseline 20-session pattern window and capped equal allocation. Compute signals after completed sessions and execute entries and close-based exits at the next actual session open. This explicitly adapts the source's execution to the supported session-market fill rule. The four inspected figures show DTW temporal alignment, manually selected template matching, historical performance curves and claimed statistics; none reveals the trading template or marked entries. All source performance claims remain unverified. The declared eight-name pilot is an instrument adaptation, not a reproduction of the all-A-share screen.

Test whether withdrawal of a large funding subsidy causes its former recipients to unwind over subsequent days. Trade INJUSD.HYPERLIQUID on 4-HOUR bars, long and short, holding at most 72 hours. This is an original, untested hypothesis. Funding signals positioning incentives rather than serving as the intended income source. [Hyperliquid documentation](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/funding) describes hourly funding payments. Market orders execute at the signal bar close, the registry equivalent to continuous-market next-bar-open execution. Protective stops trigger intrabar through platform matching. Use actual joint funding/bar coverage; the snapshot records INJ funding from 2023-05-12. No preliminary backtest was performed.

Test convergence of BTC-versus-ETH implied-minus-realized volatility premia. Buy the cheaper-premium asset's ATM straddle and sell the richer-premium asset's same-expiry straddle, matching dollar vega. Hedge each asset's delta with Bybit linear perpetuals. ETH-PERPETUAL.DERIBIT-4-HOUR-LAST-EXTERNAL is the untraded primary clock; synchronized BTC perpetual bars supply BTC realized volatility. Historical Deribit chain expansion must resolve all four actual option contracts point in time and persist their full IDs in extra_instruments and the run's market context before verification. The listed perpetuals are permanent clock and hedge instruments, not substitutes for options. Continuous-market orders execute at synchronized signal-bar closes, the registry's equivalent of next-bar-open execution. Test only actual overlapping chain, quote and hedge history. Standard-tier option fees are underlying-based: [Deribit fees](https://support.deribit.com/hc/en-us/articles/25944746248989-Fees).

Derived by the Research Lead from the publisher abstract of [Variance Decomposition and Cryptocurrency Return Prediction](https://www.cambridge.org/core/journals/journal-of-financial-and-quantitative-analysis/article/variance-decomposition-and-cryptocurrency-return-prediction/9995E58095453CB44A3BC3C9C111969F): test whether upside-dominated volatility predicts relative underperformance. The abstract attributes negative cross-sectional return predictability partly to positive variation; the full methods were not read. This is an operationalization, not a replication. Use synchronized 4-HOUR ATOM, ALGO and ICP Hyperliquid perpetual bars to buy the lowest signed-semivariance share and short the highest, only when their preceding net moves are modest relative to realized variation. Hold each paired episode for seven days. Four-hour observations are coarse proxies for the paper's intraday estimators; this transfer is explicitly unproven. Market orders fill at the signal bar's close, the registry's faithful equivalent to continuous-market next-bar-open execution. No queue, independent opposing positions, selectable same-bar policy or exchange-hosted protection is required.

Derived by the Research Lead from ChaoZhang's article, https://www.fmz.com/strategy/438771: operationalize its equity-curve gating concept on three independent Hyperliquid perpetuals using 4-HOUR bars. The supplied Pine code trades every Supertrend flip and only plots filtered performance; this adaptation explicitly implements a causal gate rather than claiming to reproduce its orders. Each instrument maintains an ungated virtual reference strategy while actual trading may pause. Market orders execute at signal-bar closes, the registry's equivalent of next-bar-open execution on continuous markets. Use available 4H history without assuming more than approximately two years. Tier-0 taker fees are 0.045% per fill, verified in [Hyperliquid's official fee schedule](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/fees).

Test whether large intrabar excursions without comparable close-to-close variance identify fragile speculative liquidity that subsequently underperforms assets with more persistent repricing. Use synchronized 4-HOUR bars across six Hyperliquid perpetuals. Buy the two lowest range-variance/close-variance scores and short the two highest for seven days. This is an original conjecture, not an established premium. Market orders execute at completed signal-bar closes, the registered equivalent of next-bar-open execution on continuous venues. Protective stops trigger intrabar through the platform. Evaluate actual available history; Hyperliquid's approximately 5000-candle limit bounds four-hour coverage. Base-tier taker fees are 0.045% each way according to [Hyperliquid fees](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/fees).