A published fast/slow moving-average crossover controlling 2× daily SPY exposure.
DEFAULT MARKETSPY
BAR INTERVAL1d
GROSS EXPOSURE2.0x
PAPER CASH$100,000.00
REALIZED P&L$+0.00
PROMOTION AUDIT
No AI qualification submitted
This strategy keeps its course or research provenance. A normal backtest snapshot cannot award the AI-generated label.
NOT TESTEDLocked evidenceRequires 100+ trades over 180+ untouched calendar days.
NOT TESTEDCost survivalMust beat exposure-matched Buy & Hold at 4 and 8 bps.
NOT TESTEDIntraday flatnessEvery position must close within its trading session.
NOT TESTEDImmutable provenanceExact algorithm, dataset, and trade-ledger SHA-256 fingerprints are mandatory.
How the strategy works
The five-day average reacts to recent price direction while the 150-day average defines the long regime. Leverage is switched on only when the fast average is above the slow one.
Concrete example
If yesterday's 5-day SPY average is 610 and its 150-day average is 575, today's SPY return is counted at 2× minus financing. A bearish crossover moves the following session to cash.
Published evidence motivates this candidate; it does not guarantee this implementation will pass our current sample.
Exact algorithm
Calculate 5-session and 150-session simple moving averages from adjusted SPY closes.
Use yesterday's completed signal to avoid look-ahead.
Take 2× daily-reset exposure while SMA(5) is above SMA(150); otherwise hold cash.
Charge 5% annual financing plus 4 bps per exposure change.
Required data: adjusted_ohlcv
Where it may work—and where it can fail
EXPECTED BEST CONDITIONS
Sustained equity trends with turning points slow enough for the crossover to exit before the deepest decline.
IMPORTANT CONTEXT
The unleveraged rule did not beat recent SPY Buy & Hold; this tested version wins the absolute-return comparison through leverage.
Main risks
Leverage magnifies gap losses
Crossovers lag rapid reversals
Sideways markets cause repeated switches
The unleveraged implementation failed the absolute-return gate
“Best conditions” is an interpretation of the strategy's mechanism, not an additional course entry rule and not a performance guarantee.
Forward paper transactions
LEDGER START2026-08-16T22:55:49.869081+00:00
PAPER VERSIONv2 · 4145c52fd8be
STATUSrunning
STATUS DETAILNo issue
Entry
Side
Entry price
Exit
Exit price
Capital basis
Gross notional
Outcome
P&L
No forward-paper transactions yet. The ledger began at deployment and never imports historical P&L.
Capital basis is allocated paper equity. Gross notional uses the exposure stored with that exact fill; legacy fills with no recorded exposure are shown as unknown. A new semantic paper generation archives the prior run and restarts the forward clock.
Archived paper versions
Read-only prior runs preserve their executable manifest, balances, and fills. They never contribute to the active clock, cash, or qualification.
The headline Buy & Hold card is the familiar 1x adjusted-price reference. A 4x strategy must also beat this 4x daily-reset reference before leverage-adjusted outperformance can be claimed. Neither comparison is frozen qualification evidence.
HITS / MISSES7 / 9
HIT RATE+43.75%
AVG HIT / MISS+36.30% / -4.42%
LONGEST MISS STREAK5
Rolling 70/30 diagnostic
This moving 70/30 split is a dashboard diagnostic, not frozen qualification evidence. It cannot award AI-qualified status.
FINAL 30% TRADES5
STRATEGY / B&H 1x+168.02% / +103.26%
FINAL 30% VS B&H+64.77%
FINAL 30% HIT RATE+80.00%
FINAL 30% DRAWDOWN+3.32%
INDEPENDENT EVIDENCE?No
Candles and execution signals
▲ BUY▼ SELLHover a candle or marker for exact values
Shows the latest 120 available bars. BUY/SELL marks strategy entries and exits whose timestamps fall in this visible window.
Historical equity curve
Compounds each encoded net trade sequentially, including the strategy's declared exposure and modeled costs. The forward ledger starts only after deployment.
Backtest specification
STATUS / REPLICATIONcompleted · published_rule_synthetic_daily_leverage