Execution Rule: Buy when Price > 200 EMA and RSI(2) < 10 (Extreme Oversold in Bullish Trend). Automatically exit on Day 5 market close or when RSI(2) > 70.
5-day holding captures the complete weekly institutional momentum cycle (Monday open to Friday close) without exposing swing trades to weekend gap risk or multi-week market drift.
5-Day Swing Indicator Benchmark Matrix
Illustrative reference benchmark across 500+ theoretical 5-day swing trade holding windows.
| Rank & Indicator Setup | Category | 5-Day Win Rate | Avg 5-Day PnL | Profit Factor | Best Market Regime | Optimal Exit Trigger | Action |
|---|
For 5-day swing trades, RSI(2) < 10 with price above the 200-day EMA produces the highest statistical expectancy. Institutional liquidity sweeps create sharp 2-3 day pullbacks that violently snap back within the 5-day holding window.
- ✔ High win-rate (65-72%)
- ✔ Fast mean-reverting velocity
- ✖ Requires strict stop-loss discipline
MACD Histogram Expansion combined with a 9/21 EMA Golden Cross works best in strong trending markets (ADX > 25). These trades sacrifice a few percentage points of win rate for much larger 5-day reward-to-risk payouts (+8% to +14%).
- ✔ Captures multi-day breakout waves
- ✔ Excellent profit factor (> 2.1)
- ✖ Vulnerable to choppy sideways chop
Bollinger Band %B < 0.05 followed by a close back inside the lower band signals capitulation exhaust. Pair with 1.5x ATR trailing stops to let winners run into Day 4 and Day 5 while cutting structural failures early.
- ✔ Clear visual risk definitions
- ✔ Low maximum drawdown
- ✖ Requires volatility normalization
5-Day Swing Trade Indicator Workbench Mechanics & Backtest Limits
How does the workbench evaluate 5-day indicator setups, and what mathematical simplifications govern its backtest metrics?
The workbench demonstrates short-term swing trading rule logic by evaluating technical indicators against an in-browser deterministic synthetic price series. For each model, the engine identifies candlestick entry conditions (such as RSI(2) dropping below 15 above an EMA baseline), enters a theoretical position at the bar's closing price, and tracks subsequent intraday highs and lows across a holding window of 2 to 10 days. An exit triggers immediately if price crosses the profit target or stop loss threshold, or at the close of the expiration session. While useful for inspecting execution order and exit rules, the analytics reflect synthetic simulation and fixed static matrix values rather than audited historical exchange fills.
The workbench relies entirely on an in-browser pseudo-random 120-bar synthetic candlestick generator seeded deterministically with seed 42, not live or historical market tick data. The benchmark table below displays hardcoded static comparison figures (such as 68.7% and 61.4% win rates across 500+ simulated windows) rather than metrics computed live from the current asset. Furthermore, the 200 EMA baseline is calculated in code using a 60-period exponential moving average proxy to fit within the 120-bar sample, and the 'ATR Volatility Sizing' toggle is visually present but does not alter the engine's fixed $10,000 position sizing calculation. Real-world execution slippage, bid-ask spreads, overnight margin requirements, and weekend gap risks are omitted.
Try a worked example
Select the 'âš¡ #2: MACD Histogram Expansion + 20 EMA' preset button while on the default NVDA asset. The backtest engine immediately recalculates across the 120 synthetic daily bars: the active model switches to MACD Histogram Expansion, the lower indicator canvas renders a histogram and signal line, and the sample counter updates to '1 Trades Backtested' with '100% Win Rate' (1 Win / 0 Losses) and '+1.76%' average return. In the trade log below, Trade #1 appears entering on 2/19 and exiting upon '5-Day Window Expired (5d)' at +1.76%. Next, adjust the 'Holding Period' slider down from 5 to 3 days; the metric updates to '1 Trades Backtested' exiting after 3 days with '+1.56%', illustrating how time-stop expiration adjusts holding duration without requiring market order executions.
Trade Window Lifecycle & Exit Priority
The engine processes daily price bars sequentially. When an entry trigger fires—such as an oversold reading or moving average cross—the model records an entry at that candle's closing price. Over the subsequent holding sessions (default 5 days), each bar's high is compared against the profit target price (Entry × (1 + target% / 100)), and its low is compared against the stop loss price (Entry × (1 - stopLoss% / 100)). Target hits and stop hits take precedence over time exits; if neither threshold is breached, the position exits at the close of the final holding day.
To prevent duplicate entries and artificial compounding on identical setups, the engine advances its scan pointer past the completed trade's exit index before evaluating the next eligible signal.
Indicator Implementations & Architectural Proxies
The workbench implements simplified JavaScript versions of classic technical indicators, including exponential moving averages (EMA), Wilder-style smoothed Relative Strength Index (RSI), and MACD line/signal pairs. Because each generated synthetic series contains only 120 daily bars, the code calculates the '200 EMA' trend filter using a 60-period moving average proxy, which requires fewer warmup bars than a standard 200-period EMA.
The comparison matrix below the chart presents static reference fixtures rather than dynamically backtested figures across historical markets. Users should distinguish between the interactive engine running in the top panel and the static illustrative figures in the benchmark matrix.
Synthetic backtest rules and accounting assumptions
Read the explanation
This saved workbench creates one hundred twenty candles with a fixed random seed, a sinusoidal trend term and symbol specific volatility. The dates begin in January twenty twenty four. These are generated prices, not historical market quotes. Indicator signals and trade statistics come from that artificial series. Use the tool to inspect its algorithm, not as evidence of actual returns or investment performance. For a hypothetical entry at one hundred, a five percent target is one hundred five and a three percent stop is ninety seven. Suppose one later candle has a high of one hundred six and a low of ninety six. Both levels are crossed, but this code checks the target first and exits at one hundred five. It does not determine which price was reached first inside that candle. The diagram shares one price axis at thirty pixels per unit. This is an illustrative example, not a selected actual trade from the generated series. The engine assigns ten thousand units of capital to every trade. A five percent gain gives five hundred; a three percent loss gives minus three hundred. In a hypothetical collection of two such wins and one such loss, win rate is sixty six point seven percent, profit factor is about three point three three, and average dollar result is two hundred thirty three point three three. The equity chart adds those results, while each new trade still uses ten thousand. The ATR sizing toggle does not change the source trade capital formula. Only a positive return is a win. A zero return is counted in the other bucket. If there are gains and no dollar losses, the displayed profit factor is forced to five, rather than mathematical infinity. The next entry search starts after the prior exit, so trades do not overlap. Fees, slippage and uncertain intraday order are not modeled. Inspect the actual generated trade export and native controls, then reset. This explainer describes saved code, not trading advice.