Strategy Card

Weekly-Daily Trend Alignment Momentum for US Large-Cap Stocks — strategy card

Spec ID: spec-multitimeframe-momentum-us-stocks-1783167355_v8 · Generated: 2026-07-10 09:52 UTC

Cluster: Momentum · Sub Cluster: Weekly-Daily Trend Alignment

One-line description

Long-only US large-cap momentum strategy that buys stocks only when daily and completed-week trend filters are simultaneously bullish. Positions are ATR-sized, capped per name, and exited on trend deterioration, trailing stop, or max holding age.

Why this trade exists

The hypothesis is that persistent equity trends are more robust when short/intermediate daily momentum is confirmed by a slower weekly trend regime. Requiring both horizons to agree attempts to avoid low-quality breakouts and whipsaw periods while retaining exposure to large-cap winners with institutional sponsorship.

The structural rationale is behavioral and flow-based: investors underreact to improving price trends, trend-following and benchmark-relative flows can reinforce moves, and large liquid stocks can sustain multi-week momentum. ATR-based sizing normalizes risk across names, while the trailing stop and trend exits aim to cut failed trends before they dominate portfolio risk.

Algorithm

[code omitted from public view]

Parameters

Param	Value	Notes
Rule type	MULTI_TIMEFRAME_TREND_MOMENTUM	Long-only trend-following momentum
Daily trend	SMA20 > SMA50, close > SMA20, SMA20 5-day slope > 0	Daily confirmation filter
Weekly trend	SMA20w > SMA50w, weekly close > SMA20w, SMA20w 4-week slope > 0	Uses completed Friday-resampled weeks only
Entry liquidity	20-day average dollar volume >= $10,000,000	Skip illiquid names
Position sizing	ATR volatility adjusted	Risk target 1% of equity per position
Position cap	10% of equity per symbol	Before leverage constraints
Max leverage	4.0x	Portfolio-level cap
Exit rules	Trend failure, 3.0x ATR trailing stop, or 126 max holding days	ATR period is 14 days
Rebalance	Daily at close	Signals and execution use close-based data
Universe	20 US stocks, top-cap filtered list	Includes AAPL, AMZN, BRK share classes, GOOG/GOOGL, MSFT, NVDA, TSLA, and others
Backtest interval	2020-01-01 to 2026-07-01	Available data period shown as 2018-01-01 to 2023-01-01 in the run context
Costs	$0.004/share commission, $1 minimum/order, max commission 0.01%, 0 bps slippage	Slippage assumption is optimistic

Look-ahead audit

#	Concern	Status
1	Weekly signal may use incomplete current week	Mitigated by use_completed_week_only=true and Friday completed-week resampling
2	Close-to-close execution can use same close as signal	Needs careful implementation; if signal and fill both occur at the same close, results may be optimistic unless orders are modeled as MOC submitted before close
3	Universe selection by current top capitalization can introduce survivorship/look-ahead bias	Requires point-in-time universe construction; provided symbol list appears static
4	Corporate actions and share-class handling	Must use split/dividend-adjusted prices and consistent identifiers for BRK-A/BRK-B, GOOG/GOOGL, etc.
5	Missing data and stale prices	Policy skips trades/signals when required inputs are missing; still audit sparse symbols and halted names
6	Transaction costs and market impact	Commissions included, but slippage is set to 0 bps, which likely understates implementation drag

Caveats / known limitations

• Static 20-name universe may overstate realism if constituents were selected with hindsight.
• Backtest context shows a data period ending 2023-01-01 despite a planned interval through 2026-07-01; confirm the actual evaluated dates before comparing results.
• Zero slippage is optimistic, especially for close-based rebalancing and high turnover.
• Same-close signal/execution needs explicit market-on-close assumptions to avoid look-ahead bias.
• Long-only momentum can suffer during sharp reversals, factor crashes, and broad equity bear markets.
• Weekly confirmation reduces noise but can delay exits and entries.
• Total trades are high for a 20-stock universe, so small cost-model errors may materially affect net results.

Results

The backtest produced a 64.29% total return with a Sharpe of 0.80, Sortino of 0.92, and Calmar of 0.59. Max drawdown was -13.57% with 12.43% volatility, indicating moderate risk-adjusted performance rather than an unusually strong standalone anomaly. The win rate was low at 36.05%, but the profit factor of 1.55 and larger average wins than losses suggest the edge came from asymmetric trend capture across 1,326 trades. Results should be interpreted cautiously given the static universe, zero-slippage assumption, and date/data-period ambiguity.

Backtest metrics snapshot

Metric	Value
Total Return	64.29%
Sharpe	0.80
Sortino	0.92
Calmar	0.59
Max Drawdown	-13.57%
Volatility	12.43%
Win Rate	36.05%
Profit Factor	1.55
Total Trades	1326
Symbols	20 (AAPL, ACON, AMZN, BETR, BRK-A, BRK-B, GOODO, GOOG, GOOGL, HOLO, +10 more)