EMA 9/21 cross + VWAP
NAS100 · 5m chart · strategy s21_ema_cross_vwap · 1 variation
Buys NAS100 on 5-minute candles when a fast average crosses above a slower one and price confirms above the day's average traded price (VWAP), then lets winners run with a wide trailing stop.
What it is about
A moving-average crossover is one of the simplest momentum signals: when the fast EMA 9 (an average of the last 9 candles, weighted to the most recent) rises above the slower EMA 21, short-term momentum has turned up.
To avoid buying into a weak market, the next candle must also close above the VWAP (volume-weighted average price: the average price paid since the trading day began at 18:00 New York, with busier candles counting more). Above the VWAP, buyers have the upper hand for the day.
The stop sits under the recent swing low, and there is no target: once a trade is 3R up (R = the amount risked), the stop follows 3R behind the best price. Many trades lose 1R; a few run far enough to pay for them.
How it trades
NAS100 at any hour of the trading day (18:00 to 16:15 New York), about 36 trades a month. Intraday only; nothing is held past 16:15.
- About 1 trade in 3 wins (35%). Most trades are 1R losses or small trailing exits; the profit comes from the runs past +3R.
- Backtest 2019-today at zero cost: 3,358 trades, +649.4R, largest drawdown about 30.7R, every year positive.
- Some years are much thinner than others: 2024 +42.0R and 2025 +42.1R against +116R to +122R in 2020, 2021 and 2023.
- Its daily results have almost nothing to do with the VWAP + EMA 9 strategy's first-two-crosses version (correlation 0.07), so the two complement each other.
- Buys only, on a market that rose for most of 2019-2026. A long falling market would hurt.
- Weaker recently: 2019-23 +465.0R vs 2024 to Oct 2026 +184.4R.
- VWAP uses tick volume, which differs by broker: on Pepperstone 5-minute data (July 2024 to Oct 2026) it made about a quarter less than on the backtest data.
- Picked from a very large search (about 380,000 settings); robust neighbours are encouraging, but only paper trading can confirm it.
Where it comes from
From a trading video Victor shared on 2026-10-18, which claimed an 80% win rate with an ADX trend filter. Testing about 380,000 settings found 27-36% wins and losses on every market for the rules as shown, and the ADX filter hurt. What held up: NAS100 buys with a swing stop of at least 15 points and wide exits (89-95% of 360 nearby settings profitable in 2019-23).
Main variation: s21-A ★ Main
The variations
Each variation is the same strategy with a different approach, trading its own paper account. Click one to see exactly what it changes.
Variations side by side
Live result of all 1 together: −2.8R (zero cost).
| Live paper account (R, zero cost) | Backtest 2018 – today (R, zero cost) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| s21-A★ MainEMA 9/21 cross + VWAP | NAS100 | 5m | −2.8R | 25 | 28% | – | +649.4R | +0.19R | 8/8 | 36.1 | 35% | −30.7R |
Main = the variation with the most backtest R per R of max drawdown (2018 – today, zero cost). It is picked on that history, so the live accounts are the real test.