Hypotheses

Mechanism: the 08:30 ET slot carries most US tier-1 macro releases: NFP, CPI, PPI, retail sales, GDP and jobless claims. These move real yields and the dollar, and gold reprices on them more directly than any crypto asset does. The first 15 minutes are dealer and algorithmic repricing. Real-money macro and CTA positioning (ETF flows, the 10:00 ET London PM fix and the US session) then continues to push in the same direction for the rest of the New York morning and afternoon. The test asks whether that slower follow-through is tradable in spot gold. No hard-coded event calendar is needed: an event is any weekday whose 08:30-08:45 ET 15-minute bar (stamped 08:45 ET, because bars are close-stamped) has a high-low range of at least k times the median range of the same slot over the previous N weekdays. A release that moves gold detects itself; quiet release days and no-release days stay below the threshold. DST is handled by converting to America/New_York with the platform session calendar. Direction comes from close location in the event bar. Close in the top fraction q of the range means long; close in the bottom q means short; a close in the middle means no trade (ambiguous reaction). Execution: SPOTMETAL is a session venue, so the decision on the 08:45 ET bar close fills at the next bar's open (next_open_fill, supported) with a market order. The exit is a market order at a fixed New York time (exit_time_et), always at or before 16:00 ET. The position therefore never crosses the 17:00 NY overnight-financing cut and pays no financing. Protective stop: an intrabar reduce-only stop-market at the event bar's opposite extreme (below the event-bar low for longs, above the high for shorts), matched inside the bar by the platform; no exchange-resting order is required. At most one trade per day. Novelty against prior art: the crypto versions failed. 'US Macro-Release-Hour Repricing Continuation' on the AAVE USD-M perp (fee_edge, -0.24%/trade) and the two ETH macro-event continuation tests (fee_edge/overfit) traded an asset with no first-order link to US rates, paid ~0.10% round trip, and used a calendar of only 3 event types. This test moves the mechanism to the asset whose price is directly a function of US real rates and the dollar. Costs are ~0.015% round trip (6-7x cheaper), and the data-slot detector is calendar-free. Existing spot-gold hypotheses in the corpus are Asian-range / London-open breakouts, an Asia-session drift, a previous-day-level OCO breakout and daily macro-trend ideas; none conditions on the US 08:30 ET release slot. Diversity: SPOTMETAL holds 0.2% of experiments; the strategy is long/short and short-horizon (15m), against a corpus that is 52.7% BINANCE and 77.8% long-only. Fee arithmetic: in the 2003-05 to 2007-07 catalog slice (the Dukascopy backfill is still progressing, so this is all the history currently on disk), event days at k=2 with a decisive close-location show a mean absolute 08:45->16:00 ET move of ~0.8%. If continuation holds 55-58% of the time with a symmetric payoff, the signed capture is ~0.12-0.20% per trade, 8-13x the ~0.015% round-trip spread. Disclosure: that same small early slice (58 events, 2003-2007) gave a NEGATIVE signed mean at k=2 (-0.29%). Direction is pre-registered as continuation from the economic argument and is fixed, not tuned. If full history confirms fading, the hypothesis is falsified, not flipped. Firing rate: ~14 events/yr at k=2 in the early slice, more in the higher-volatility post-2008 regime. Expect 250-450 trades over 2003-2026 once backfilled. Risk control route (a): position size = risk_pct x equity / stop distance, where stop distance = |entry - opposite event-bar extreme|. This loses 0.75% of equity at the stop, with notional capped at max_leverage x equity (fixed 4x, below the 20x venue cap).

Test equal-base-quantity, opposite DOGE perpetual positions on synchronized 1-DAY bars. Require persistent funding dispersion, hold for weeks, and reject an unfavorable entry basis. Use all actual overlapping bar and funding history. Market orders execute at the signal bar close, the capability registry equivalent of next-bar-open execution on continuous venues. Separate instruments support opposing positions without hedge mode; fills are sequential, not guaranteed atomic. Funding mechanics are documented by [Hyperliquid](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/funding) and [Bybit](https://www.bybit.com/en/help-center/article/Funding-fee-calculation?category=cd60af6303161fd598). Current runner.py credits funding despite a stale statement in data_specification.md; use existing engine accounting without adding strategy-level credits.

Derived by the Research Lead from OKX Learn's discussion of Dogecoin's sensitivity to sentiment: test whether prices moving against contemporaneous news reveal demand or supply stronger than the public narrative, followed by another 24–72 hours of repricing. The supplied local article and [linked page](https://www.okx.com/learn/does-dogecoin-still-have-a-future) were read. The article supplies no trading rules; this is a Research Lead operationalization. Trade DOGEUSD.HYPERLIQUID long and short on 4-HOUR bars, using DOGEUSDT.BINANCE_SPOT for price confirmation only. News sentiment is the platform's DOGE-mapped, ingestion-time feed with MACRO merged; it does not directly measure tweets or social engagement. Market entries execute at signal-bar close, the registry equivalent of next-bar-open on continuous venues. Protective stops trigger intrabar through platform matching. Evaluate only the common sentiment-covered interval beginning 2026-08-13, currently about 60 days; earlier price bars supply warm-up only. Predict 6–12 qualifying episodes per month. Evidence may remain inconclusive because the sentiment history is short.

Test whether Hyperliquid funding changing from disagreement to agreement with a persistent Binance funding regime predicts subsequent directional price adjustment. Trade AAVE, UNI and LTC Hyperliquid perps independently on synchronized 4-HOUR bars; matching Binance perps provide signals only. This is an original hypothesis. Funding is a positioning proxy, not funding-income arbitrage. Hyperliquid settles funding hourly with an eight-hour formula, as described in [Hyperliquid's documentation](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/funding); use the platform's documented normalized eight-hour rates without multiplying already-normalized values. Market orders execute at the signal-bar close, the registry equivalent of next-bar-open execution on continuous venues. Baselines: persistence_days=3, excess_threshold=0.00005, hold_bars=18. Predicted firing rate: 10–25 events per coin-year. Risk route (b): signal/time exits with capped unlevered sizing. Evaluate all genuinely overlapping available history.

Derived by the Research Lead from QuantStart's State Space Models and the Kalman Filter: https://www.quantstart.com/articles/State-Space-Models-and-the-Kalman-Filter/. The supplied local full text explains causal filtering but supplies no trading rules or profitability evidence. Test first-turn reversals of large ZEC/DASH relative-price deviations using synchronized 4-HOUR bars, trading ZECUSD.HYPERLIQUID against DASHUSDT.BYBIT. Exact catalog directories exist; filenames indicate common history beginning April 2025, subject to continuity verification. Market orders at the signal close implement the capability registry's continuous-market equivalent of next-bar-open execution. No atomic fills, hedge mode or exchange-resting protection are required. Baselines: observation_bars=180, state_noise_ratio=0.001, entry_z=2.0, hold_bars=30. Predicted frequency is 15–30 paired episodes annually; no performance pre-study was run.

Test whether sustained Binance spot buying against perpetual selling, or spot selling against perpetual buying, predicts multi-day adjustment toward the spot flow. Trade Hyperliquid perpetuals on completed 4-HOUR decision bars; Binance spot and USD-M instruments supply actual aggressor-signed trade-tick signals only. Baseline accumulation is 48 hours and holding time is 72 hours. Expected frequency is 12–30 entries per asset/year, an untested prediction. Market orders execute at signal-bar close, the registry equivalent of next-bar-open execution on continuous venues. Protective stops trigger intrabar through platform matching. Use all common available history; do not assume Hyperliquid's approximately 2.3-year 4H API coverage can be extended retrospectively. Input definitions follow [Binance public-data documentation](https://github.com/binance/binance-public-data); fees follow [Hyperliquid official documentation](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/fees). These sources document inputs and costs, not evidence for this edge.

Derived by the Research Lead from the safe-haven literature (Baur & Lucey 2010, 'Is Gold a Hedge or a Safe Haven?'; Baur & McDermott 2010): gold earns positive returns in the days right after extreme stock-market losses, and that haven bid fades and partly reverses after about 15 trading days. Operationalization: the SIGNAL comes from SPY.USEQ daily bars (16:00 ET close). The TRADE is XAUUSD.SPOTMETAL spot gold on 1-DAY bars (trading day closes 17:00 New York). Both are session venues. A SPY shock known at 16:00 ET is acted on at gold's next bar OPEN (18:00 NY session reopen), the platform's supported next_open_fill rule. No lookahead: the SPY bar is stamped at 16:00 ET, before the 17:00 NY gold bar close that triggers the decision. On a SPY 'shock-down' session (SPY close-to-close return <= -shock_z times its trailing 60-session return std, with the std computed from the sessions BEFORE the shock), buy spot gold at the next gold open and hold for long_hold_sessions gold trading days. If reversal_leg = on, then when the long hold ends, flip short for reversal_hold_sessions days to capture the documented haven unwind. A new shock during a hold restarts the long clock (it does not add size). A new shock during the reversal leg closes the short and goes long again. Why this instrument: XAUUSD.SPOTMETAL has 9 experiments in the corpus and data back to 2003, so this spends unused data. SPY is used only as a signal source and is never traded. Cross-venue and long-short both fill under-represented quota buckets. It is mechanically different from the prior gold failures: GLD momentum confluence, Asia drift, the London breakout and PAXG premium were all gold-only price patterns, while this is a cross-asset event trigger. Data note: the 1-DAY spot gold catalog is currently backfilled from 2003 to 2007, and the rest is collected on demand from Dukascopy. The backtest needs the full 2003-2026 span. Risk control is route (a) plus (b): signal/time exits, a fixed 50% of equity notional (unlevered), and an intrabar protective stop at 4% adverse from entry (2% of equity at the stop). The stop triggers inside the bar in the platform's simulated venue and live mirrors those fills; no exchange-resting order is required.

Single-instrument, long/short regime strategy on spot gold (XAUUSD.SPOTMETAL, 1-DAY bars, Dukascopy mid prices, history from 2003; the signal starts in mid-2008 when GVZ begins). Signal: the gold variance risk premium VRP_t = GVZ_{t-1} - RV21_t, where GVZ is the CBOE Gold ETF Volatility Index (30-day implied vol of GLD options, in vol points) and RV21 is the annualized close-to-close standard deviation of the last rv_lookback daily log returns of XAUUSD.SPOTMETAL x sqrt(252) x 100. Each session the strategy computes the percentile rank of VRP_t within its own trailing rank_window values. It goes LONG when the rank reaches entry_rank, meaning option buyers are paying an unusually rich premium over realized gold risk. It goes SHORT when the rank falls to 1 - entry_rank, meaning implied vol is cheap relative to realized turbulence. Each side is held with hysteresis until the rank crosses back through exit_band around the median. This is the variance-risk-premium return predictor of Bollerslev, Tauchen & Zhou (2009, RFS, 'Expected Stock Returns and Variance Risk Premia') applied to gold's own options market. It is also a deliberate CROSS-ASSET test of the one VRP mechanism that has survived in this factory: 'ETH Variance-Risk-Premium Regime Long on Deribit's Own Linear Perp' (promoted, Sharpe 0.81, avg trade +2.66%, 36 trades). Novelty vs the corpus: no prior hypothesis uses GVZ or any gold implied-vol input. All nine prior SPOTMETAL runs are intraday session-breakout/drift rules, OHLCV-only reversion, or the BTC/XAU relative-momentum rotation, so none of them is a vol-premium regime trade. SPOTMETAL is 0.2% of the corpus and BTC/Binance is over-concentrated, so this spends an unused series. Execution: SPOTMETAL is a session venue, so a decision taken on daily bar N's close (17:00 New York) fills at bar N+1's OPEN (the 18:00 session reopen), which the platform supports natively (next_open_fill). No stop orders are needed. Risk-control route (b): signal/regime exit with capped sizing. Position notional is fixed at 100% of equity (unlevered, never above 100%), and long and short do not overlap because the account is netting and a flip closes before reversing. Overnight financing is charged by the engine's funding ledger on every 17:00 roll (long pays fed funds + 2.5%/yr, short receives fed funds - 2.5%/yr, Wednesday x3), and the edge arithmetic below includes it. Report long-leg and short-leg PnL separately so the analyst can tell whether the long side alone (the side with the survivor precedent) carries the result. DATA: GVZ is NOT yet an injectable supplementary key (only VIX is mirrored from FRED). FRED publishes it as GVZCLS (daily from 2008-06-03), and the existing FRED macro collector should add it under the key US_GVZCLS with the same availability-time stamping as US_VIXCLS: a value appears at 00:00 UTC the day after its date. A daily decision at the 17:00 NY gold close therefore reads the PREVIOUS US session's GVZ close through supp_as_of('US_GVZCLS'), never a same-day value. See data_needs.

Single-instrument long/short trade on spot gold (XAUUSD.SPOTMETAL, 1-DAY bars, 2003-2026). It is driven ONLY by gold's two textbook macro drivers, never by gold's own price trend. (1) The US 10-year TIPS real yield (FRED DFII10, platform key US_REAL10Y): gold pays no coupon, so the real yield is its opportunity cost. (2) The US dollar, proxied by EUR/USD (FRED DEXUSEU, key US_EURUSD; EUR is ~58% of DXY): gold is priced in dollars. The bet is that trends in these drivers PERSIST (bond and FX time-series momentum, Moskowitz-Ooi-Pedersen 2012), and that gold's price response spreads over weeks through slow ETF and allocator flows. A lagged macro trend should therefore predict gold's next few weeks. Signal per decision bar: s_r = -(R_t - R_{t-N}) / sd(N-obs changes in R over the trailing 252 obs); s_d = (ln E_t - ln E_{t-N}) / sd(N-obs log changes of E over the trailing 252 obs). score = mean of the active components. signal_mode is both / real_yield_only / dollar_only; the single-driver modes are the matched controls that show which driver carries information. Both series are read ONLY via supp_as_of / supp_window. The platform keys daily FRED series at 00:00 UTC the day AFTER their date, so a decision at the gold daily close (17:00 New York) uses the previous US day's yields and FX: strictly point-in-time, one day stale. Counts are in published observations, not calendar days (FRED skips holidays). Execution: SPOTMETAL is a session venue, so a decision at a daily bar's close fills at the NEXT session's open (18:00 NY Sunday-Thursday) under next_open_fill (supported). Weekends and the 17:00-18:00 break are closed market, not data gaps. Novelty vs. the corpus: SPOTMETAL has 9 experiments, all intraday session/London-open breakouts, BTC/XAU relative momentum or a PAXG premium. The failed GLD 20/120 dual-horizon confluence used GOLD'S OWN momentum. No prior hypothesis uses real yields or the dollar as an outside signal for gold; a knowledge search for TIPS, DFII10, real yield and copper found no matching strategy. The macro feeds have full history (FRED from 2003), so this is not a short-lookback design. Overnight financing is modelled: long pays fed funds + 2.5%/yr, short receives fed funds - 2.5%/yr (Wednesday x3). Longs therefore carry a real drag in high-rate years (about 0.02%/day in 2023-24), and the edge must clear it. Trade-rate estimate: N=20 and entry_z=0.5 give roughly 12-18 entries/flips per year, about 300 round trips over 23 years, with a median hold of about 2-4 weeks. Per-trade arithmetic: gold's 15-session move has sd ~3.5-4%. A 53-55% directional hit rate with symmetric payoff captures ~0.5-0.7% gross per trade. Subtract financing on longs (~0.15-0.3% over a 15-day hold at 2023-24 rates, much less in 2009-2021) and ~0.015% round-trip spread, giving ~0.35-0.5% net, more than 20x the round-trip cost. Known risk stated upfront: from 2022 gold decoupled from real yields (central-bank buying), so the real-yield component may have pointed the wrong way in 2022-2025. The dollar_only control and the regime attribution will show it.

Derived by the Research Lead from the FMZ essay on falsifiable trading decisions and avoiding additions to losing positions: https://www.fmz.com/bbs-topic/973. The supplied local article was read; it supplies a philosophy of explicit invalidation, not these entry rules. This operationalization tests whether a completed spot breakout followed by a successful later retest of its frozen boundary predicts multi-day continuation. Use synchronized 4-HOUR bars for three independent spot/perpetual pairs. Trade only Hyperliquid perps, both long and short; Binance spot instruments are signal sources. No trade occurs on the initial breakout. Market orders execute at the completed confirmation bar close, the capability registry equivalent of next-bar-open execution on continuous venues. Intrabar protective stops are matched by the platform, with no exchange-hosting requirement. Baselines: range_bars=30, retest_window_bars=12, retest_band_atr=0.35. Predicted frequency is 12-24 completed campaigns per asset/year, not a measured rate. Use all actual overlapping history, with no assumption that Hyperliquid 4H history can be backfilled indefinitely.

Event-driven calendar strategy on SPY.USEQ daily bars, built on the pre-FOMC announcement drift documented by Lucca & Moench (2015, Journal of Finance, 'The Pre-FOMC Announcement Drift'): from 1994 to 2011 the S&P 500 earned roughly 49 bp on average in the 24 hours before scheduled FOMC statements, which is a large share of the total equity premium over that period. The commonly cited explanation is resolution of macro uncertainty and dealer risk-bearing ahead of the announcement. The strategy holds SPY only around the 8 SCHEDULED FOMC statement days per year and is in cash otherwise, so it is exposed about 2-4% of sessions. The FOMC calendar is public, known-in-advance history. The developer hard-codes the scheduled statement dates (federalreserve.gov FOMC historical calendars, 1994 to present, plus the published 2026 schedule) as a constant list of dates in the strategy; for two-day meetings the statement date is the second day. Unscheduled or emergency actions (e.g. 2001-09-17, 2008-10-08, 2020-03-03, 2020-03-15) are EXCLUDED because they were not known in advance. No supplementary feed is needed (L189). Execution: USEQ is a session venue, so a decision taken on the close of daily bar N fills at the OPEN of session N+1 (supported next_open_fill). Market-on-close is unsupported, so the literature's exact 14:00-to-14:00 window cannot be replicated on daily bars; the contract instead holds from the open of the session before the statement day to either the open of the statement day or the open of the session after it. Each choice is a tunable. Novelty: the knowledge base has no FOMC/macro-event-calendar hypothesis (searched 'FOMC'), and this is not momentum, mean reversion or trend. The fact that it is flat ~97% of the time makes it uncorrelated with the deployed portfolio. SPY is used deliberately because the documented effect is on the S&P 500, and SPY has 30+ years of daily bars covering about 260 scheduled events. Fee arithmetic: USEQ round trip is about 0.03-0.05% (spread plus impact, zero commission). The literature drift is about 0.49% per event pre-2011 and is reported weaker since (Kurov, Wolfe & Gilbert 2021 find it attenuated after 2015). I expect our open-to-open window to capture 0.15-0.30% per event on average over the full sample, which is 3-6x the round-trip cost. Firing rate: 8 events/yr x ~32 years = ~260 trades; the sandbox's 20k-bar budget covers the whole daily history.

Derived by the Research Lead from Thomas DeMark's Sequential using artificial intelligence, https://www.mql5.com/en/articles/2773: test the idea of waiting for a variable-length directional setup to terminate before fading it. This is an OHLCV-only operationalization, not a reproduction of its neural networks, proprietary inputs or full TD Sequential. Trade XLM, ATOM, AAVE and UNI perpetuals independently on closed 4-HOUR bars, long and short. Baselines: comparison_lag=4, min_run_bars=9, min_run_move_pct=6.0, run_volume_ratio=1.25, hold_bars=18. Predict 15–30 entries per instrument-year; this firing rate is unmeasured. Use each instrument's available history after warm-up, never inventing pre-listing bars. Hyperliquid's 5000-candle limit provides approximately 2.3 years at 4H. Market orders execute at signal-bar close, the capability registry's equivalent to next-bar-open execution on a continuous venue. Protective stops trigger intrabar through platform matching. Risk route (a): 0.5% equity risk per position at an initial 2-ATR stop, capped at 15% equity notional per instrument and 60% total gross exposure, leverage 1.0.

Derived by the Research Lead from 同花顺 SuperMind《振幅大于1、周线macd在零轴之上、昨日股价大于250日均线》：保留文章公式中的三个筛选条件，将没有退出和资金管理规则的选股信号补全为可交易策略。分类为部分规则，采用 distilled。原文Python示例与公式不一致：示例使用区间平均振幅、周线ma30和均线交叉，且所示半年数据不足以计算250日均线；本实验明确以“同花顺指标公式代码参考”为准，不宣称逐行移植该示例。研究选择六只固定A股、日线决策、次交易日开盘成交、条件失效退出及资金管理。只使用已完成周线，不使用形成中的本周数据。二维码来源读取失败，没有用于制定规则。文件系统只读，未写入指定outbox；本回复提供完整任务JSON。

Port 善's explicit SuperTrend trading rules from the FMZ article, using 15-minute BCHUSDT.OKX linear perpetual bars instead of the historical BTC/USD inverse quarterly contract. Preserve period 45, factor 3, both trading directions, and exit-before-reentry behavior. Adapt the source's repeated aggressive two-tick limit orders to ordinary market orders at the signal bar's close; this preserves directional intent but does not reproduce its polling, cancellation or partial-fill timing. Replace the fixed one-contract amount with 20% of current equity in position notional, without leverage amplification. The source has no separate stop-loss, take-profit or time stop; do not add them. All six supplied figures were inspected: the cover and GitHub image add no trading rules; the comparison charts show BTC daily SuperTrend 10/3 overlays around the March 2020 decline; the parameter screenshot confirms pd=45, factor=3, Amount=1, TradeInterval=2000ms, PriceTick=0.5 and Symbol=quarter; the performance screenshot shows an unverified historical BTC/USD quarterly backtest. The trading experiment uses the article's explicit 15m and 45/3 settings, not the illustrative daily-chart settings or GitHub issue's 8/2.1 settings.

Derived by the Research Lead from Angeris, Chitra, Evans and Lorig's A primer on perpetuals: operationalize the linear-payoff replication argument as long AVAX spot on Binance and short an equal AVAX quantity on Hyperliquid when historically observed funding offers sufficient compensation for trading costs. For phi(S)=S, Section 3's replication holds one underlying unit per contract unit and its idealized funding expression is zero. Actual exchange perpetuals have different contractual mechanics; their positive funding is a candidate carry opportunity, not an arbitrage guaranteed by the theorem. The authors specify no exchange strategy, entry threshold or holding period. The market, 4-hour decision interval, persistence filter, cost hurdle and exits below are Research Lead choices. Use signal-close market execution on these continuous venues, the registry's equivalent to next-bar-open execution. This is one applicable operationalization; it does not pretend that listed linear perpetuals reproduce the paper's logarithmic or power contracts.

Faithful port of the paper's MACD&SAR combination: a Parabolic SAR flip against closing price is tradable only when MACD agrees. Preserve MACD 12/26/9, SAR acceleration 0.02 and maximum acceleration 0.2 from Section 2.5. SAR is an indicator used at daily close, not a resting protective stop. Use DIA, QQQ and SPY daily with independent one-third-initial-capital sleeves; invest 95% of available sleeve equity in whole shares on entry, no shorts, no pyramiding and leverage 1. These universe and sizing choices adapt the source's constituent experiments. Platform split-adjusted OHLCV and next-session-open fills replace unspecified price-adjustment and execution details. Add no target, protective stop or time limit. Treat the paper's performance claims as unverified.

Derived by the Research Lead from Pat Tong Chio's paper: test the volume-, typical-price- and volatility-weighted MACD derivative on DIA, QQQ and SPY daily bars. Preserve the published 12/26/9 periods, 0.1 bandwidth and both printed signal inequalities. Equation 4.2-6 uses SVWMA rather than LVWMA; this experiment deliberately implements that literal expression, without claiming to resolve the author's intent. Population rather than sample standard deviation is an explicitly chosen convention. Replace the unspecified constituent universe with the three ETFs explicitly named in Section 3.2.2, using split-adjusted platform OHLCV. Allocate independent, non-transferable sleeves of one-third initial capital, with 95% of each sleeve's available equity invested on entry. Whole shares, no shorts, leverage 1. Orders decided at session close fill at the next session open. No added stop or target: risk is bounded by sleeve allocation rather than a guaranteed loss limit. Source performance claims remain unverified.

Faithful port of Table 4's MACD&RSI trading rules, retaining MACD 12/26/9, Wilder RSI 14, entry threshold 35 and exit threshold 70. Every one of the six RSI observations must satisfy the relevant threshold; this is not an any-observation test or an RSI crossing rule. Trade DIA, QQQ and SPY daily as independent one-third-capital sleeves, investing 95% of each sleeve when signaled, rounded down to whole shares. The ETF universe and reduced sleeve exposure are disclosed adaptations of the paper's constituent-level, all-capital experiments. Use split-adjusted prices, no dividends in PnL, no shorts and leverage 1. Close-derived decisions execute at the next session open. Preserve signal-only exits without adding a stop, target or holding-period cap. Claimed source results are unverified.

This is the article's cash-and-carry funding harvest, changed where our own data shows the original no longer works. For each coin in a pool, the strategy holds a long spot position plus an equal-notional short perpetual. That pair has almost no price exposure, so its P&L is the funding the short collects, minus fees and minus drift in the basis. Pre-proposal check on our supplementary data: the article's same-exchange version (Binance spot long + Binance USD-M short) still worked in 2020-2024. Across 507 Binance spot/perp twins, episodes won 73-91% of the time with a median net gain of 2-4% after a 0.30% round trip. In 2025-26 it broke down to median net about 0% and a 50% win rate, and on the 12 most liquid non-BTC Binance pairs a 7-day trailing funding above 0.06%/day NEVER fired after 2024. Binance funding on majors is now arbitraged down to near the 0.01%/8h interest floor. Hyperliquid perps still pay materially positive funding, because the venue is structurally long-crowded and retail-heavy with hourly funding. So the short leg moves to HYPERLIQUID and the hedge stays on the most liquid spot book (BINANCE_SPOT). Pool of 18 coins, BTC excluded for corpus-concentration reasons: ETH, SOL, XRP, DOGE, SUI, ENA, NEAR, WLD, ARB, LINK, AVAX, LTC, AAVE, ADA, BNB, SEI, WIF, CRV. Selection rule: HL perps that have a Binance USDT spot twin, are among the most liquid on HL, and are not new listings or meme coins under a year old (the article's 'avoid new coins and 妖币' rule). A coin is eligible only on days when both legs have a bar. This pool is a hand choice and is declared as such (lesson 151); the Binance-spot-twin requirement is its only economic justification. A daily-bar carry simulation on this pool (funding only, basis ignored) gave: 2023 36 episodes, mean net +6.8%; 2024 58, +4.9%; 2025 49, +1.5% (90% win); 2026 YTD 5 episodes, about +0.1%. Duty cycle was about 34% of coin-days, with a typical hold of 20-55 days. It is an honest carry premium that shrinks as the regime cools, and it is NOT a price prediction.

Separately registered lower-risk replication of the existing DOT/OKX adaptation of the FMZ JavaScript state machine. Retain completed daily bars, strict SMA 5/15 crossings, signal-close market execution and exit-only opposite transitions. Change only sizing_risk_fraction from 0.01 to 0.005 and max_notional_fraction from 0.20 to 0.10. Target entry notional becomes equity × min(0.10, 0.005/0.05) × 1 = 10% of equity, versus the prior 20%. A 5% adverse instrument move therefore corresponds to approximately 0.5% of entry equity before costs; the reference move is a sizing assumption, not a stop or guaranteed loss limit. The state machine depends on the first eligible crossing: ordinary alternating crossings can repeatedly enter the initial direction and use opposite crossings only to exit. Balanced directional exposure is not claimed. This experiment tests the revised risk contract; reducing exposure does not establish an economic edge. Source figures and full text were not supplied in this revision session, so their earlier inspection claims are not independently repeated.