NQ: Sweep, change of character, fair value gap
Sweep of a session level, then a change of character, then entry on the fair value gap. Backtested on real CME NQ 1-minute bars.
Short answer
We backtested the Sweep, change of character, fair value gap on NQ (Nasdaq-100 futures (NQ and MNQ)) using real CME 1-minute bars from 2 Jan 2023 to 18 Oct 2026. It took 399 trades, won 34.6% of them and made -$4,156 after costs, an average of -0.03R per trade. It did not clear the bar: it failed 7 of 8 tests, including "Average R of +0.10 or more". Against a random-walk copy of the same data it scored 0.07R better, so there is no clear edge over chance. It trades 12:30 to 20:30 IST in summer or 13:30 to 21:30 IST in winter (03:00 to 11:00 New York time).
Strategy setup
Times in your time zone: EST (America/New York), New York times beside them. Change it in your profile.
Real days from the data
Wed, 16 Sept 2026, short. In at 05:05 EST (05:05 New York), out at 05:18 EST (05:18 New York). 10-minute candles, 4 MNQ, $146 at risk.
- Levels
- Asia high, Asia low, yesterday's high, yesterday's low, overnight high, overnight low03:00 to 11:00 EST (03:00 to 11:00 New York)
- Setup
- On 5-minute bars: a sweep of a level, then a close through the lowest low (highest high) of the last 6 bars within 90 minutes
- Entry
- Limit order at the midpoint of the latest fair value gap in that move, working for 60 minutes
- Stop
- One tick beyond the sweep's extreme
- Target
- 2R, twice the distance to the stop
- Time exit
- Flat by this time at the latest11:00 EST (11:00 New York)
- Size
- Whole MNQ micros, so the stop risks at most $150 (maximum 10). Skipped if one micro would risk more
- Frequency
- One trade a day
One tick lost on stop and market fills, a target needs price to trade a tick through it, the stop counts first when one bar touches both, and a target can't be hit in the bar a limit order fills. Commission is a placeholder $0.50 per micro per side, not Tradeify's actual rate.
Does it clear the bar?
Doesn't meet the barFixed before any strategy was run. Every check must hold after costs. Average R is -0.03 against -0.10 on random-walk copies of the same days (±0.03): no better than chance.
- At least 300 trades399
- Average R of +0.10 or more-0.03
- t-statistic of 3 or more-0.4
- Beats a random walk by 0.10R+0.07R over -0.10R
- Profitable in 3 of 4 years1 of 4
- Positive out-of-sample (2025 on)-0.13R
- Profit factor of 1.2 or more0.88
- Max drawdown within $1,000-$4,954
Net profit
-$4,156
after $1,301 costs
Trades
399
on 40.6% of days
Win rate
34.6%
avg win $213, avg loss -$128
Average R
-0.03
95% range -0.15 to +0.11
Profit factor
0.88
expectancy -$10 a trade
Max drawdown
-$4,954
vs $1,000 allowed on a $25K account
Equity curve
Running net profit, by quarter
By year
Net profit, and trades taken
How trades ended
Net profit by exit type
Days with a trade
Share of each year's trading days on which the strategy traded, and why the other days had no trade.
- 202346%
- 202442%
- 202538%
- 202635%
| Reason for no trade | Days | Share of all days |
|---|---|---|
| no setup in the window | 547 | 55.7% |
| one micro risks more than the budget | 36 | 3.7% |
Breakdowns
Nothing is tuned on the out-of-sample years: 2023 to 2024 is the in-sample period, 2025 on is out-of-sample.
| Period | Trades | Win rate | Net | Avg R | Profit factor | Max drawdown |
|---|---|---|---|---|---|---|
| In-sample 2023 to 2024 | 228 | 36.4% | -$518 | +0.05 | 0.97 | -$2,551 |
| Out-of-sample 2025 to Oct 2026 | 171 | 32.2% | -$3,638 | -0.13 | 0.75 | -$4,780 |
| By year | Trades | Win rate | Net | Avg R | Profit factor | Max drawdown |
|---|---|---|---|---|---|---|
| 2023 | 119 | 35.3% | -$764 | +0.01 | 0.92 | -$2,543 |
| 2024 | 109 | 37.6% | $246 | +0.09 | 1.03 | -$2,025 |
| 2025 | 98 | 37.8% | -$147 | +0.04 | 0.98 | -$2,255 |
| 2026 | 73 | 24.7% | -$3,491 | -0.36 | 0.48 | -$3,491 |
| By side | Trades | Win rate | Net | Avg R | Profit factor | Max drawdown |
|---|---|---|---|---|---|---|
| Long | 158 | 36.1% | -$737 | +0.01 | 0.94 | -$2,992 |
| Short | 241 | 33.6% | -$3,419 | -0.05 | 0.84 | -$3,600 |
Rule variations
Small changes to the rules, to check the result is not fragile. They are never used to choose the rules.
| Variation | Trades | Win rate | Net | Avg R | Profit factor | Max drawdown | Days skipped |
|---|---|---|---|---|---|---|---|
| Primary rules | 399 | 34.6% | -$4,156 | -0.03 | 0.88 | -$4,954 | - |
| Target 1R | 399 | 50.6% | -$1,548 | +0.01 | 0.94 | -$3,362 | 583 |
| Target 1.5R | 399 | 42.1% | -$1,311 | +0.02 | 0.96 | -$3,721 | 583 |
| Target 3R | 399 | 27.6% | -$6,657 | -0.08 | 0.82 | -$7,470 | 583 |
| Gap order valid 30 min | 332 | 35.2% | -$2,686 | -0.01 | 0.90 | -$3,804 | 650 |
| Gap order valid 90 min | 442 | 35.1% | -$4,010 | -0.01 | 0.89 | -$4,822 | 540 |
| Structure 3 bars | 420 | 32.6% | -$6,574 | -0.08 | 0.82 | -$8,045 | 562 |
| Risk budget $250 | 429 | 34.3% | -$6,003 | -0.04 | 0.90 | -$8,119 | 553 |
Tradeify $25K evaluation
Target $1,500, trailing drawdown $1,000, daily loss $600. An attempt starts on each of 982 trading days in turn.
- Passed 116
- Failed on drawdown 844
- Unfinished 22
12.1% of finished attempts passed, in a median of 82.5 trading days. Attempts that start on consecutive days overlap, and with no edge a simulated evaluation passes or fails mostly by luck.
Hypothetical results from a backtest on CME NQ futures, 2 Jan 2023 to 18 Oct 2026, run 19 October 2026. 104 strategies and variations were tried across the Strategy section, so the best few would look good by chance. Past performance does not predict future results, fills and costs are assumed, and an average of -0.03R with a 95% range of -0.15 to +0.11R can't be told apart from zero. Education, not trading advice.
More strategies
- NQ: 9:30 opening range breakoutTrade the first break of the 15-minute range after the New York open. Stop on the other side, 2R target, flat by 11:30.
- NQ: London breakout of the Asia rangeEnter when price breaks the Asia high or low during London. Stop on the other side of the range.
- NQ: Asia range sweep and reversalFade a sweep of the Asia high or low that closes back inside the range.
- NQ: Previous-day high or lowSweep-and-reject at yesterday's high or low.
- GC: Sweep, change of character, fair value gapSweep of a session level, then a change of character, then entry on the fair value gap.
Trading a prop firm account?
Track each challenge's rules and today's dollar limits, with your sessions in your time zone. Free.