Key results
- Total return
- -30.3%
- Buy & hold: +2.1%
- Annual return (CAGR)
- -3.5%
- Buy & hold: +0.2%
- Max drawdown
- -33.5%
- Buy & hold: -27.6%
- Win rate
- 33.89%
- Profit factor
- 0.8
- Trades
- 1,009
- Longest losing streak
- 14 trades
- Period
- Oct 2, 2016 to Oct 1, 2026
- 10 years
- Market
- GBP/USD
- 15-minute
Equity curve vs buy & hold
Returns year by year
| Year | Return | Trades |
|---|---|---|
| 2016 | -2% | 25 |
| 2017 | -1.5% | 103 |
| 2018 | -4.7% | 96 |
| 2019 | -0.1% | 92 |
| 2020 | -4.7% | 108 |
| 2021 | -3.2% | 97 |
| 2022 | -2.4% | 96 |
| 2023 | -4.7% | 104 |
| 2024 | -7.1% | 101 |
| 2025 | -3.8% | 109 |
| 2026 | -1.3% | 78 |
Calendar years, compounded from monthly results. The first and last years can be partial.
Kill zones are one of the most recognisable ideas in ICT (Inner Circle Trader) material: rather than trading all day, you only take setups in the windows when the big sessions open and volume arrives. A popular way to use them, described in guides such as this ICT Silver Bullet explainer, is to wait for a fair value gap inside the kill zone and trade the retest in the direction of the move. We tested that core entry on GBP/USD, one of the pairs most associated with London and New York session trading, inside the New York AM kill zone, over ten years.
The rules we tested
- Market: GBP/USD, 15-minute candles.
- Period: 2 October 2016 to 1 October 2026, ten years.
- Kill zone: entries only inside the New York AM kill zone, 12:00 to 15:00 UTC. No other session is allowed.
- Long entry: price returns into a bullish fair value gap on the 15-minute chart (first touch), inside the kill zone, and the 15-minute close is above the 50-period EMA.
- Short entry: price returns into a bearish fair value gap (first touch), inside the kill zone, and the close is below the 50-period EMA.
- Exit: a fixed stop-loss of 30 pips and a fixed take-profit of 60 pips, a 1:2 risk-reward. 28 trades closed on a strategy signal instead.
- Direction: long and short.
- Costs: 0.02% commission per side, full position size on each trade.
This is a reduced reading of the Silver Bullet. The full sequence (liquidity sweep, then market structure shift, then fair value gap) was not tested here; only the fair value gap entry inside the New York AM kill zone, with a trend filter.
What the backtest shows
The strategy lost 30% over ten years, about -3.5% a year, on 1,009 trades. GBP/USD itself ended almost flat: buy and hold returned +2%. The strategy's maximum drawdown reached about 33%, deeper than the pair's 28%.
About 34% of trades won: 647 hit the stop, 334 hit the target and 28 closed on a signal. The average win was about 0.42% of the account and the average loss about 0.27%. The profit factor was 0.80, so the strategy won back only about 80 cents for every dollar it lost. The longest losing streak was 14 trades in a row.
Every calendar year finished negative, and the losses were steady rather than concentrated. 2024 was the worst at -7%, with 27 winners out of 101 trades. The least bad full year, 2019, ended almost flat (-0.07%). Every other full year lost between 1.5% and 4.7%.
Just above break-even before costs, then costs
With a 60-pip target and a 30-pip stop, a strategy needs to win one trade in three just to break even before fees. This one won 33.89%. Then each round trip paid 0.04% of the position, and 1,009 round trips add up to about 40% of position value paid in commissions over the decade. That was enough to turn a flat edge into a steady loss.
Why it doesn't work here
The kill zone did its job as a filter: it kept the strategy to about 100 trades a year, every one of them during the New York morning. But fewer trades did not mean better trades. The win rate landed almost exactly on the break-even line of a 1:2 bracket, which means the fair value gap retest, even inside the kill zone and with the trend, carried no measurable edge on this pair.
Costs then decided the result. A 30-pip stop on GBP/USD is small, so 0.04% per round trip is a real share of every trade. After costs, the average win was only about 1.6 times the average loss instead of 2 times, and at that ratio you need to win nearly two trades in five to break even. The strategy won about one in three.
None of this settles whether the full Silver Bullet sequence works. It shows that the fair value gap retest alone, inside the New York AM kill zone, did not carry an edge on this pair.
What you could test next
- Add the missing steps: require a liquidity sweep and a change of character before the fair value gap, as the full model describes.
- Move the fair value gap to the 1-hour chart and keep the 15-minute chart for timing, so only larger imbalances qualify.
- Widen the stop and target together, for example 50 and 100 pips, so commissions weigh less on each trade.
- Compare kill zones: run the same rules in the London open window, or on EUR/USD and the Nasdaq 100.
Methodology and assumptions
- Starting capital
- 10,000
- Position size
- 100% of equity
- Commission
- 0.02% per side
- Data window
- Oct 2, 2016 to Oct 1, 2026
- Run date
- October 2, 2026
- No repainting: every signal is computed on closed candles only, so the backtest never acts on a price a trader could not have seen yet.
- Past performance does not predict future results. A backtest is a historical simulation, not a forecast.
Frequently asked questions
What is an ICT kill zone?
In Inner Circle Trader (ICT) teaching, kill zones are specific windows of the trading day, such as the London open or the New York morning, when institutional activity and volatility are expected to be highest. Setups are only taken inside those windows. This test used the New York AM kill zone, 12:00 to 15:00 UTC.
Is this backtest the full ICT Silver Bullet?
No. The Silver Bullet as usually taught is a sequence: a liquidity sweep, then a market structure shift, then a fair value gap entry inside the kill zone. We tested the part described as the entry itself, a fair value gap retest inside the New York AM kill zone with a trend filter, without the sweep and structure-shift steps.
What win rate did the kill zone FVG strategy get?
About 34% over 1,009 trades. With a 60-pip target against a 30-pip stop, you need about 33.3% just to break even before costs, so the strategy sat barely above break-even before commissions, and below it after.
Why did the kill zone strategy lose money?
Because the bracket is small next to the cost of trading. Each round trip costs 0.04% of the position, a real slice of a 30-pip stop. After costs, the average win (about 0.42% of the account) was only about 1.6 times the average loss (about 0.27%), not the 2 times the 60 and 30 pip bracket suggests, and a 34% win rate cannot carry that.
Reproduce it, then change it
Every number on this page comes from the Backtrex engine. Rebuild the strategy in the app, then change the market, the timeframe or a parameter and see whether the result holds.
Ready to backtest your strategies?
Create your free account and take the guided tour: a real 10-year backtest in 3 minutes, no coding, no credit card.
Create your account in 30 seconds and run your first backtest. No credit card.
Other backtests
- Strategy
- -60%
- Buy & hold
- +2.1%
- Strategy
- -65.6%
- Buy & hold
- +2%
- Strategy
- -15.3%
- Buy & hold
- +2%
- Strategy
- +26.9%
- Buy & hold
- +528.3%