Tested by BacktestingMax
Higher highs and higher lows (market structure) backtest
Does trading higher highs and higher lows make money?
Mostly no. We backtested the standard rules on 10 markets, each on 15-minute, 1-hour and 4-hour charts, from 2020 to 2025, after trading costs. 9 out of 30 backtests made money. Only 3 made money in both 2020–2023 and 2024–2025. On gold, the account grew 306% on 1-hour charts and 40% on 4-hour charts. On the DAX 4-hour chart, the account grew 62%. On 15-minute charts, only 1 out of 10 markets made money. In an uptrend, each peak and each dip on the chart is higher than the one before. The standard rules buy after a new higher low, when a candle closes above the high of the candle that made that low. The stop-loss goes at the higher low, and the trade closes when a lower low forms. In a downtrend, the strategy sells after a new lower high instead. We also tested the idea behind the strategy: after a higher low, price goes on up. In 56.8% of 31,661 trades, price got back to the last peak before the stop-loss (for a sell, to the last dip). A price moving at random would do that 57.4% of the time. So a higher low did not make price more likely to keep rising.
30 backtests 2020–2025 Trading costs included Tested 9 November 2026
Where it worked 30 backtests- Made money
- 9 of 30 tests, with the standard rules
- Made money in both periods
- 3 2020–2023 and again 2024–2025
- Best backtest
- +306% Gold, 1-hour chart: account growth 2020–2025
- With a take-profit at twice the risk
- 4 tests made money in both periods, compared with 3 with the standard rules
Made money in both periods
- Gold, 1-hour
- Gold, 4-hour
- DAX, 4-hour
The results
Results on every market
Each box shows one backtest of the higher highs and higher lows strategy with the standard rules, from 2020 to 2025, after trading costs: the % result, and a small chart of the account balance. The solid line is the account balance and the dashed line is the starting balance. Green results made money; red results lost money. Under each result: whether it made money in both 2020–2023 and 2024–2025, or passed all four of our checks (see "How a test passes" further down).
| Market | 15-minute | 1-hour | 4-hour |
|---|---|---|---|
| EUR/USD | −99% Lost money. Account down 90% by Jan 2023 | −71% Lost money | −8% Lost money |
| GBP/USD | −99% Lost money. Account down 90% by Dec 2021 | +45% Made money overall, lost in one period | −11% Lost money |
| USD/JPY | −81% Lost money. Account down 90% by Sept 2023 | +86% Made money overall, lost in one period | +29% Made money overall, lost in one period |
| Gold | −97% Lost money. Account down 90% by Oct 2022 | +306% Made money in both periods | +40% Made money in both periods |
| Silver | −99% Lost money. Account down 90% by Mar 2021 | −80% Lost money | −5% Lost money |
| Nasdaq 100 | −66% Lost money | −12% Lost money | −3% Lost money |
| S&P 500 | −88% Lost money | +26% Made money overall, lost in one period | −5% Lost money |
| Dow Jones | −88% Lost money | +3% Made money overall, lost in one period | −9% Lost money |
| DAX | +25% Made money overall, lost in one period | −44% Lost money | +62% Passed all four checks |
| FTSE 100 | −99% Lost money. Account down 90% by Oct 2022 | −80% Lost money | −27% Lost money |
All the numbers, test by test
| Market | Chart | Trades | Win rate | Profit factor | Result | Max drawdown | 2020–2023 | 2024–2025 | Monte Carlo drawdown | Trading costs |
|---|---|---|---|---|---|---|---|---|---|---|
| EUR/USD | 15-minute | 3,585 | 28.2% | 0.80 | −98.7% | −98.8% | −93.4% | −80.2% | −99.2% | $6,285 |
| EUR/USD | 1-hour | 929 | 28.1% | 0.86 | −71.1% | −79.3% | −36.4% | −54.5% | −81.7% | $3,309 |
| EUR/USD | 4-hour | 228 | 31.6% | 0.94 | −7.8% | −20.2% | +3.5% | −10.9% | −34.9% | $366 |
| GBP/USD | 15-minute | 2,753 | 25.5% | 0.78 | −99.3% | −99.5% | −99.1% | −28.0% | −99.6% | $5,723 |
| GBP/USD | 1-hour | 913 | 30.8% | 1.07 | +44.8% | −42.2% | −5.6% | +53.4% | −59.1% | $4,708 |
| GBP/USD | 4-hour | 214 | 34.1% | 0.91 | −10.5% | −19.7% | −8.4% | −2.3% | −32.3% | $352 |
| USD/JPY | 15-minute | 3,662 | 29.9% | 0.89 | −81.2% | −94.0% | −86.8% | +42.2% | −95.0% | $10,374 |
| USD/JPY | 1-hour | 919 | 33.2% | 1.09 | +85.9% | −32.6% | +87.9% | −1.0% | −51.0% | $5,074 |
| USD/JPY | 4-hour | 229 | 28.8% | 1.14 | +28.5% | −21.2% | +55.0% | −17.1% | −34.0% | $469 |
| Gold | 15-minute | 2,774 | 27.5% | 0.86 | −96.9% | −98.0% | −94.6% | −42.9% | −98.5% | $9,195 |
| Gold | 1-hour | 850 | 32.6% | 1.24 | +306.1% | −44.2% | +130.6% | +76.1% | −43.4% | $7,968 |
| Gold | 4-hour | 234 | 34.6% | 1.27 | +40.3% | −12.6% | +31.0% | +7.1% | −24.3% | $482 |
| Silver | 15-minute | 1,253 | 22.2% | 0.59 | −98.7% | −98.7% | −98.5% | −9.5% | −99.0% | $8,098 |
| Silver | 1-hour | 852 | 25.6% | 0.72 | −80.2% | −81.7% | −73.8% | −24.5% | −86.4% | $4,522 |
| Silver | 4-hour | 231 | 30.7% | 0.96 | −4.7% | −26.1% | −7.0% | +2.5% | −32.5% | $962 |
| Nasdaq 100 | 15-minute | 3,467 | 28.4% | 0.95 | −66.0% | −76.7% | −58.5% | −18.1% | −92.5% | $10,813 |
| Nasdaq 100 | 1-hour | 871 | 26.6% | 0.98 | −12.1% | −46.8% | −16.1% | +4.8% | −67.4% | $2,151 |
| Nasdaq 100 | 4-hour | 226 | 36.3% | 0.97 | −2.9% | −20.5% | −1.1% | −1.8% | −28.0% | $178 |
| S&P 500 | 15-minute | 3,416 | 28.9% | 0.94 | −87.6% | −90.2% | −65.4% | −64.1% | −96.3% | $12,589 |
| S&P 500 | 1-hour | 885 | 29.3% | 1.03 | +26.0% | −40.8% | +33.8% | −5.8% | −59.6% | $3,334 |
| S&P 500 | 4-hour | 227 | 34.8% | 0.96 | −4.9% | −19.2% | +0.1% | −5.0% | −32.4% | $225 |
| Dow Jones | 15-minute | 3,449 | 27.9% | 0.96 | −88.1% | −95.6% | −65.7% | −65.3% | −96.3% | $13,802 |
| Dow Jones | 1-hour | 880 | 28.5% | 1.01 | +3.2% | −54.4% | −20.0% | +28.9% | −61.7% | $1,610 |
| Dow Jones | 4-hour | 235 | 30.6% | 0.92 | −8.9% | −26.8% | −1.8% | −7.2% | −34.8% | $145 |
| DAX | 15-minute | 3,487 | 31.3% | 1.01 | +25.2% | −66.0% | −43.2% | +120.6% | −84.6% | $10,587 |
| DAX | 1-hour | 889 | 29.7% | 0.91 | −44.2% | −56.1% | −23.2% | −27.3% | −69.9% | $1,226 |
| DAX | 4-hour | 216 | 37.5% | 1.48 | +62.2% | −8.3% | +17.0% | +38.6% | −20.5% | $161 |
| FTSE 100 | 15-minute | 3,248 | 27.3% | 0.84 | −99.0% | −99.3% | −95.8% | −75.2% | −99.5% | $6,378 |
| FTSE 100 | 1-hour | 908 | 28.2% | 0.77 | −79.9% | −85.8% | −61.1% | −48.2% | −85.7% | $1,879 |
| FTSE 100 | 4-hour | 241 | 29.5% | 0.76 | −27.1% | −36.4% | −22.9% | −5.5% | −41.8% | $300 |
Profit factor is the money won divided by the money lost. Above 1 means the test made money. Max drawdown is the biggest fall from the highest balance. 2020–2023 and 2024–2025 show the result in each period on its own. Monte Carlo drawdown: we shuffled the trades into a random order 1,000 times; in 95% of those runs, the worst drawdown was this number or less. Result is the change in the account, with the risk of each trade set to 1% of the account at that time.
The account balance
The best test and the middle test
What we found
What the 30 backtests show
With the standard rules, 3 backtests made money in both 2020–2023 and 2024–2025: gold on 1-hour and 4-hour charts, and the DAX on 4-hour charts. Only the DAX on 4-hour charts also passed all four of our checks. On this page, R is the amount a trade is set to risk: 1% of the account balance at the time. In those 3 backtests, buys made +237.6R and sells only +19.8R. Gold and the DAX both rose a lot from 2020 to 2025. On 15-minute charts, the higher low is close to the entry, so the spread is a big share of each trade. The average 15-minute trade made +0.006R before costs and lost 0.071R after.
Does a higher low make price more likely to keep rising? To check, we took the 31,661 trades that had not yet passed the last swing point when they opened. For a buy, that is the last swing high, the most recent peak. For a sell, it is the last swing low. In 56.8% of these trades, price reached that swing point before it hit the stop-loss. A price moving at random would do that 57.4% of the time, from the same entry and stop-loss. The two numbers were close on all three charts. So in this test, the higher low gave no advantage. A take-profit at the last swing high would not have helped either. On the same trades, the average trade would have lost 0.016R before trading costs.
Many smart money (SMC) traders only trade in the direction of the higher-timeframe trend: the trend on a longer chart. We tested that too, with the same swing-point rule on the longer chart. We used the 4-hour trend for 15-minute trades, and the daily trend for 1-hour and 4-hour trades. This filter kept about a third of the trades: 14,868 instead of 42,271. In 2024–2025, 14 out of 30 backtests made money with it, compared with 9 out of 30 with the standard rules. In 2020–2023, 7 out of 30 did, compared with 8 out of 30. So the filter did not clearly help. On 4-hour charts, the filter kept only 61 to 87 trades per backtest, too few to judge.
Does the trend continue?
Price reached the last swing point about as often as a random price would
For each trade, we checked which came first after the entry: the last swing high (for a sell, the last swing low) or the stop-loss. We used 1-minute prices. Standard rules, all 10 markets, 2020–2025.
| Chart | Trades checked | Reached the swing point first | A random price would |
|---|---|---|---|
| 15-minute | 23,638 | 57.0% | 57.7% |
| 1-hour | 6,551 | 54.3% | 54.9% |
| 4-hour | 1,472 | 64.9% | 63.0% |
| All three charts | 31,661 | 56.8% | 57.4% |
'A random price would' is the chance that a price moving up and down at random reaches the swing point before the stop-loss. It is the distance from the stop-loss to the entry, divided by the distance from the stop-loss to the swing point. For example, say the stop-loss is 10 points below the entry and the swing high 5 points above it. Then a random price reaches the swing high first 10 times out of 15: 67%. In 61.1% of these trades, the swing point was closer to the entry than the stop-loss was. That is why both numbers are above 50%. We worked this out for each trade, from the chart price at the entry without the spread, and show the average. When both prices came in the same minute (22 trades), we counted the stop-loss. We left out the 10,594 trades that opened with price already past the last swing point. On 4-hour charts, buys reached the swing high more often than a random price would (70.7%, compared with 65.4%, in 788 trades). Sells did so less often (58.2%, compared with 60.3%). Most of these markets rose from 2020 to 2025, which helps buys and hurts sells. Our random price does not rise or fall on average.
How the trades ended
6 trades in 10 hit the stop-loss at the higher low
- Hit the stop-loss
- 61.6% 26,026 trades. On average, each lost 1.02R: a little more than 1R, because of trading costs
- Closed when a lower low formed
- 38.4% 16,233 trades. On average, each made +1.50R, and 74.5% of them won
Standard rules, all 30 backtests: 42,271 trades, after trading costs. 12 more trades were still open at the end of 2025 and closed then. A lower low can close a buy only after price has made a newer higher low above the stop-loss. So the trade usually closes above its entry, and most of these trades won. On 4-hour charts, fewer trades hit the stop-loss: 55.4%. Most trades lose about 1R, so the strategy needs its fewer, bigger winners to pay for them.
Daily charts, more years
On daily charts, 7 out of 10 markets made money, but not much
Extra backtests with the standard rules, not part of the 30 above. Daily charts give only about 6 trades a year, so these backtests use more years. They run from 2007 to 2025 for currency pairs, gold and silver, and from 2014 to 2025 for the stock indices. After trading costs.
| Market | Trades | Whole period | Before 2020 | 2020–2025 |
|---|---|---|---|---|
| EUR/USD | 128 | +14.4% | +5.1% | +8.9% |
| GBP/USD | 122 | +4.8% | −6.8% | +12.5% |
| USD/JPY | 127 | +15.1% | +15.9% | −0.7% |
| Gold | 108 | +19.8% | +13.4% | +5.6% |
| Silver | 114 | +6.1% | +13.5% | −6.5% |
| Nasdaq 100 | 70 | +14.9% | +2.3% | +12.4% |
| S&P 500 | 79 | −8.5% | −11.1% | +2.9% |
| Dow Jones | 68 | +4.6% | +10.3% | −5.2% |
| DAX | 78 | −27.0% | −9.3% | −19.5% |
| FTSE 100 | 84 | −27.8% | −18.1% | −11.8% |
'Before 2020' is 2007–2019 for currency pairs, gold and silver, and 2014–2019 for the indices. The best result, gold's +19.8% in 19 years, is about 1% a year. No daily backtest had more than 128 trades, and the indices had 68 to 84, fewer than the 100 our checks ask for. So luck can explain much of these results. The three changes did not make daily charts much better. With a take-profit at twice the risk, 6 out of 10 made money, and the best was the Nasdaq 100 (+25.2%). With the exit at the first close below the last higher low, 6 out of 10 made money, and the best was gold (+25.1%). With the higher-timeframe trend (the weekly chart's trend for daily trades), each backtest had only 16 to 51 trades.
The strategy
What the higher highs and higher lows strategy is
In an uptrend, each peak on the chart is higher than the one before it: a higher high. Each dip is also higher than the one before it: a higher low. A downtrend has lower highs and lower lows. This way of defining a trend is one of the oldest. It comes from Dow theory, which goes back to Charles Dow's writing around 1900. Traders who use smart money concepts (SMC) or the ICT (Inner Circle Trader) method call this pattern of highs and lows market structure. The standard rules we tested: in an uptrend, buy after a new higher low, put the stop-loss at that low, and close the trade when a lower low forms. To find the peaks and dips, we used swing points. A swing high is a candle whose high is above the highs of the 3 candles on each side of it. A swing low is a candle whose low is below the lows of the 3 candles on each side of it.
- Swing points
- 3 candles on each side. We did not test other swing sizes
- Stop-loss
- At the higher low itself (for a sell, at the lower high), with no extra distance. It does not move
- Take-profit
- None with the standard rules: a lower low closes the trade. One of the changes under 'Can the higher highs and higher lows strategy be improved?' adds a take-profit
- Risk per trade
- Each trade is sized so that, if price goes from the signal candle's close to the stop-loss, the account loses 1% of its balance at the time. The account started at $10,000. The trade opens at the next candle's open, and its size is rounded down to a size the market allows. So the real risk can be a little more or less than 1%
- Trading costs
- The spread: the gap between the buy price and the sell price. Currency pairs: 0.8 pips on EUR/USD, 1.2 on GBP/USD and 0.9 on USD/JPY (a pip is 0.0001 on EUR/USD and GBP/USD, and 0.01 on USD/JPY). Metals: $0.30 on gold and $0.020 on silver. Stock indices: 1.5 points on the Nasdaq 100, 0.4 on the S&P 500, 2.0 on the Dow Jones, and 1.0 on the DAX and the FTSE 100. For silver, the Nasdaq 100, the Dow Jones, the DAX and the FTSE 100, we used wider spreads than our backtest engine's defaults, to match a typical retail broker. Our test treats chart prices as the middle between the buy price and the sell price. So a trade pays half the spread when it opens, and half when it closes at the stop-loss or at the market price. In the take-profit version, a take-profit fills at its own price and pays no spread. With a real broker, those winning trades would pay about half a spread more. There is no commission
- Skipped trades
- Sometimes a trade that risks 1% would be smaller than the smallest trade a market allows. The smallest trades are 1,000 units of a currency pair, 1 ounce of gold, 50 ounces of silver, or 0.1 of an index contract. The test then skips that trade. With the standard rules, the test skipped 99 trades on 1-hour and 4-hour charts. 85 of them were on silver, after its 1-hour account had lost more than half. On 15-minute charts, it skipped 155 trades while the account held more than $1,000. It skipped 7,347 more after accounts had lost 90%, because 1% of what was left was too small. Every backtest kept running after a 90% loss
- Rounded trade sizes
- Trade sizes are rounded down to a size step the market allows, so some trades risked less than 1%. On 1-hour and 4-hour charts, 1,587 trades (14%) risked less than 0.9%, most of them on silver and the stock indices
- Largest trade size
- A very close stop-loss can need a bigger trade than our backtest engine allows. The limit is 30 times the account balance on currency pairs, and 20 times on gold, silver and the indices. Then the trade uses the biggest size allowed and risks less than 1%. This happened in 189 trades, all on 15-minute charts
- Overnight (swap) fees
- Not included. Brokers charge (or sometimes pay) a fee for each night a trade stays open. Half of the trades closed within 3.6 hours on 15-minute charts, 14 hours on 1-hour charts and 3.4 days on 4-hour charts. On 4-hour charts, 75% of the trades stayed open longer than a day. On daily charts, half stayed open about 3 weeks or longer. So with a real broker, the 4-hour and daily results could be higher or lower
- Price data
- CFD prices: the prices a broker quotes for a CFD (contract for difference), the product most brokers outside the US offer for trading these markets. We built the candles from 1-minute prices. 4-hour and daily candles start at 17:00 New York time. The 1-minute prices also show whether the stop-loss or the take-profit came first inside a candle. Candles with no price movement (holidays in the data) are never swing points or signal candles. Neither are the tiny extra candles that a few minutes of prices after the Friday close make in some of our data
The rules we tested
- Find the swing points. A swing high is a candle whose high is higher than the highs of the 3 candles before it and the 3 candles after it. A swing low is a candle whose low is lower than the lows of the 3 candles on each side. A swing point is known only when the 3rd candle after it closes. That candle confirms it.
- Uptrend: the last swing high is higher than the one before it (a higher high). The last swing low is also higher than the one before it (a higher low). Downtrend: both are lower. If one rises and the other falls, there is no trend and no trade.
- Buy setup: a new higher low in an uptrend. Starting with the candle that confirms the higher low, wait for the first candle that closes above the high of the swing-low candle (the candle that made the higher low). That candle is the signal candle. Buy at the next candle's open.
- Put the stop-loss at the higher low itself: the low of the swing-low candle. The stop-loss does not move.
- Close the trade at the next candle's open when a lower low forms: a new swing low that is lower than the swing low before it. This can happen only after price has made a newer higher low above the stop-loss, and then a swing low below that one. There is no take-profit.
- Sells follow the same rules in a downtrend. After a new lower high, sell after the first close below the low of the swing-high candle. The stop-loss goes at the lower high, and the trade closes when a higher high forms.
- A buy setup is cancelled if price falls back to the higher low before the signal candle, if a newer swing low replaces it, or if the uptrend ends. Only one trade is open at a time. If the signal candle comes while a trade is open, that setup is not traded.
- Size every trade so that, if the stop-loss is hit, the trade loses 1% of the account.
The method
How we tested it
The test
- Markets
- EUR/USD, GBP/USD, USD/JPY, Gold, Silver, Nasdaq 100, S&P 500, Dow Jones, DAX and FTSE 100
- Charts
- 15-minute, 1-hour and 4-hour candles
- Dates
- January 2020 to December 2025
- Account
- $10,000, risking 1% of the account on each trade
- Trading costs
- Included. Every trade pays that market's normal trading cost: the spread, which is the gap between the buy price and the sell price.
How a test passes
We call a test passed only if it meets all four checks. A strategy can make money in one backtest and still fail one of these checks.
- At least 100 trades. With fewer trades, a good result could be luck.
- It made money from 2020 to 2023 and again from 2024 to 2025. This shows whether the strategy kept working in the newer years, not only in the older ones.
- A profit factor (the money won divided by the money lost) of at least 1.2 from 2024 to 2025, after trading costs.
- In 95% of 1,000 random shuffles of the trades, the account must not fall more than 30% from its highest point.
Improvements
Can the higher highs and higher lows strategy be improved?
We tried three changes that traders often make, each as its own version on the same 30 backtests. The first adds a take-profit at twice the risk. The second closes the trade earlier, at the first close below the last higher low. The third adds a higher-timeframe trend filter. If you pick the best version on the same years you judge it on, it will often look better than it really is. So we picked the version that made money in the most backtests in 2020–2023 only. Then we checked it on 2024–2025, the years we did not use to pick it.
| Version | Made money 2020–2023years used to pick the best version | Made money 2024–2025years not used to pick it | Made money in both periods | Passed all four checks |
|---|---|---|---|---|
| Standard rules The rules under 'The rules we tested': the stop-loss at the higher low, no take-profit, and the trade closes when a lower low forms. | 8 of 30middle test −21.4% | 9 of 30middle test −6.5% | 3 | 1 |
| With a take-profit at twice the riskBest on 2020–2023 Close the trade when it has made twice what it would lose at the stop-loss (2R), measured from the close of the signal candle. A lower low no longer closes the trade. | 12 of 30middle test −27.4% | 11 of 30middle test −5.2% | 4 | 2 |
| Exit at the first close below the last higher low The trade also closes at the next candle's open after a candle closes below the most recent higher low (for a sell, above the most recent lower high). The trade does not wait for a lower low to form. | 10 of 30middle test −24.6% | 11 of 30middle test −8.7% | 5 | 1 |
| With the higher-timeframe trend Only buy when a longer chart is also in an uptrend by the same swing-point rule, and only sell when it is in a downtrend. The longer chart is the 4-hour chart for 15-minute trades, and the daily chart for 1-hour and 4-hour trades. | 7 of 30middle test −18.5% | 14 of 30middle test −1.4% | 3 | 2 |
Not by much. In 2020–2023, the years we used to pick, the take-profit at twice the risk (2R) made money in the most backtests: 12 out of 30. The standard rules made money in 8 out of 30. In 2024–2025, the years we kept for the check, the gap was smaller: 11 out of 30, compared with 9 out of 30. With the 2R take-profit, 4 backtests made money in both periods (3 with the standard rules), and 2 passed all four checks (1 with the standard rules). But a fixed take-profit closes the biggest winning trades early. With the standard rules, gold's 10 best trades on 1-hour charts made $32,342, more than that backtest's whole profit of $30,607. With the 2R take-profit, that backtest made +126% instead of +306%. The exit at the first close below the last higher low made money in both periods in 5 backtests, the most of the four versions. The higher-timeframe trend made money in 14 out of 30 backtests in 2024–2025, the most of any version, but in only 7 out of 30 in 2020–2023. With every version, most of the 30 backtests lost money from 2020 to 2025.
Passed all four checks with a take-profit at twice the risk
- Nasdaq 100, 4-hour
- DAX, 4-hour
Questions
Questions traders ask
Does trading higher highs and higher lows (market structure) work?
Mostly no, in our backtest. We tested buying after a new higher low in an uptrend, with the stop-loss at that low, and selling after a new lower high in a downtrend. We ran it on 10 markets and 3 timeframes from 2020 to 2025, after trading costs. 9 out of 30 backtests made money. Only 3 made money in both 2020–2023 and 2024–2025: gold on 1-hour and 4-hour charts, and the DAX on 4-hour charts. On 15-minute charts, only 1 out of 10 markets made money.
Is a trend more likely to continue after a higher low?
Not in our test. After a new higher low, price got back to the last swing high (the most recent peak) before the stop-loss in 56.8% of 31,661 trades. Sells were checked the same way, after a lower high. A price moving at random would reach the swing point first 57.4% of the time, from the same entry and stop-loss. The two numbers were close on 15-minute, 1-hour and 4-hour charts. Before trading costs, the average trade made almost nothing: 1.4% of the amount it risked.
What is the best timeframe for trading higher highs and higher lows?
No timeframe worked on most markets in our test. 1-hour and 4-hour charts did better than 15-minute charts: 8 out of 20 backtests made money, compared with 1 out of 10. But even there, 12 out of 20 lost money. On 15-minute charts, the stop-loss at the higher low is close to the entry, so trading costs take a big share of each trade. On daily charts, 7 out of 10 markets made money, but slowly. The best was gold, with +19.8% from 2007 to 2025, about 1% a year.
Should you only trade with the higher-timeframe trend?
Our test does not show that it helps. We took trades only when a longer chart showed the same trend, by the same swing-point rule. That was the 4-hour chart for 15-minute trades, and the daily chart for 1-hour and 4-hour trades. In 2024–2025, 14 out of 30 backtests made money this way, compared with 9 out of 30 without the filter. In 2020–2023, 7 out of 30 did, compared with 8 out of 30. The filter also cut the number of trades to about a third.
Keep exploring
Related strategies we tested
ICT & smart money Change of character (CHoCH) and market structure shift (MSS) backtest Made money in 7 of 40 tests, 5 of them in both periods. Best: USD/JPY, 1-hour, +66% ICT & smart money Order block strategy backtest Made money in 5 of 30 tests, 2 of them in both periods. Best: USD/JPY, 4-hour, +29% ICT & smart money ICT New York killzone backtest on NQ and ES futures Made money in 4 of 10 tests, 1 of them in both periods. Best: MGC (Micro Gold), 15-minute, +36% ICT & smart money ICT Power of 3 (AMD) backtest on NQ, ES and gold futures Made money in 4 of 10 tests, 2 of them in both periods. Best: YM (Dow futures), 15-minute, +46% ICT & smart money ICT Silver Bullet backtest on NQ and ES futures Made money in 0 of 10 tests, 0 of them in both periods. Best: MGC (Micro Gold), 5-minute, −2% ICT & smart money Fair value gap (FVG) retest strategy backtest Made money in 5 of 24 tests, 2 of them in both periods. Best: S&P 500, 15-minute, +124%The strategy code we ran
This is the exact code from the BacktestingMax strategy engine. Paste it into the strategy editor in the app to run it on any market.
class HigherHighsHigherLows(Strategy):
"""Higher highs and higher lows: trade with the trend that the swing points show (Dow
theory, and "market structure" in smart money concepts), plus three common changes.
Swing points. A swing high is a candle whose high is higher than the highs of the 3
candles before it and the 3 candles after it, so it is known only 3 candles later. A
swing low is the mirror image.
Uptrend: the last two swing highs rise (a higher high) and the last two swing lows rise
(a higher low). Downtrend: both fall.
Standard (all switches off):
1. A new swing low that is higher than the swing low before it, while the last two
swing highs also rise, is a buy setup: a higher low in an uptrend.
2. Buy at the next candle's open after the first candle, from the one that confirms
the swing low onward, that closes above the high of the swing-low candle.
3. Stop-loss at the higher low: the low of the swing-low candle.
4. Close the trade at the next candle's open when a lower low forms: a new swing low
below the swing low before it. No take-profit.
5. Sells mirror it: a lower high in a downtrend, sell after the first close below the
low of the swing-high candle, stop-loss at the lower high, close the trade when a
higher high forms.
6. A setup ends when price reaches its swing point, when a newer swing point of the
same kind is confirmed, or when the trend no longer holds (a buy setup ends when a
lower high is confirmed). One trade at a time: a setup that triggers while a trade
is open is dropped.
7. Size every trade so that the stop-loss costs 1% of the account, measured from the
close of the signal candle.
Skipped candles: flat candles (high = low: no price moved, e.g. a holiday in the data)
and candles that open on a Friday at 17:00 New York time or later (a few minutes of
prices after the weekly close in some index data). They are never swing points or
signals.
The changes, one at a time:
target_r = 2 take profit at 2R, where R is the distance from the signal candle's
close to the stop-loss; a lower low (higher high) no longer closes it
trail = 1 also close the trade at the next open after a candle closes below the
most recent higher low (for a sell: above the most recent lower high),
without waiting for a lower low to form
htf = 1 only buy when the higher chart is in an uptrend by the same rule (its
last two swing highs and its last two swing lows rise), only sell when
it is in a downtrend. Higher chart: M15 -> H4, H1 -> D1, H4 -> D1,
D1 -> W1. A higher-chart candle is skipped when every chart candle in
it was.
"""
params = {"swing_bars": Param(type=int, default=3, min=1, max=20),
"target_r": Param(type=float, default=0.0, min=0.0, max=10.0),
"trail": Param(type=int, default=0, min=0, max=1),
"htf": Param(type=int, default=0, min=0, max=1)}
lookback = 5
higher_chart = {"M5": "H1", "M15": "H4", "M30": "H4", "H1": "D1", "H4": "D1", "D1": "W1"}
def init(self, ctx):
self.ny = ctx.clock("America/New_York")
self.chart = {"win": [], "highs": [], "lows": []} # swing points on the chart
self.buy_setup = None # a higher low in an uptrend: {"stop", "trigger"}
self.sell_setup = None # a lower high in a downtrend
self.exit_level = None # trail version: the most recent higher low (lower high)
self.view = None
if ctx.params.htf == 1:
self.view = ctx.tf(self.higher_chart[ctx.timeframe])
self.higher = {"win": [], "highs": [], "lows": []}
self.view_seen = -1
self.view_ms = self.view.timeframe_seconds * 1000
self.counted = [] # open times of the last chart candles that were not skipped
@staticmethod
def add_candle(s, n, k, high, low):
"""Add one candle to a chart's swing points. Returns the swing high and the swing low
it confirms (the candle n candles back), each as a dict or None. "level" is the swing
price; "trigger" is the other end of that candle (its high for a swing low)."""
win = s["win"]
win.append((k, high, low))
if len(win) > 2 * n + 1:
win.pop(0)
if len(win) < 2 * n + 1:
return None, None
ck, ch, cl = win[n]
others = win[:n] + win[n + 1:]
sh = {"bar": ck, "level": ch, "trigger": cl} if all(ch > x[1] for x in others) else None
sl = {"bar": ck, "level": cl, "trigger": ch} if all(cl < x[2] for x in others) else None
if sh is not None:
s["highs"] = (s["highs"] + [sh])[-2:]
if sl is not None:
s["lows"] = (s["lows"] + [sl])[-2:]
return sh, sl
@staticmethod
def trend(s):
"""+1 when the last two swing highs and the last two swing lows both rise, -1 when
both fall, 0 otherwise."""
hs, ls = s["highs"], s["lows"]
if len(hs) < 2 or len(ls) < 2:
return 0
if hs[1]["level"] > hs[0]["level"] and ls[1]["level"] > ls[0]["level"]:
return 1
if hs[1]["level"] < hs[0]["level"] and ls[1]["level"] < ls[0]["level"]:
return -1
return 0
def on_bar(self, ctx):
p = ctx.params
n = p.swing_bars
h, l, c = ctx.high[0], ctx.low[0], ctx.close[0]
after_close = (self.ny.weekday == 4 and self.ny.minute_of_day >= 17 * 60) or self.ny.weekday == 5
counts = h > l and not after_close
if self.view is not None:
self.read_higher_chart(ctx, n, counts)
if not counts:
return
s = self.chart
sh, sl = self.add_candle(s, n, ctx.bar_index, h, l)
hs, ls = s["highs"], s["lows"]
lower_low = sl is not None and len(ls) == 2 and ls[1]["level"] < ls[0]["level"]
higher_high = sh is not None and len(hs) == 2 and hs[1]["level"] > hs[0]["level"]
trend = self.trend(s)
# The open trade: a lower low (higher high) ends a buy (sell), unless the version
# uses a fixed take-profit instead.
pos = ctx.position
if pos is not None:
if pos.is_long:
if sl is not None and self.exit_level is not None and sl["level"] > self.exit_level:
self.exit_level = sl["level"] # a new higher low
end = lower_low or (p.trail == 1 and self.exit_level is not None and c < self.exit_level)
else:
if sh is not None and self.exit_level is not None and sh["level"] < self.exit_level:
self.exit_level = sh["level"] # a new lower high
end = higher_high or (p.trail == 1 and self.exit_level is not None and c > self.exit_level)
if p.target_r <= 0 and end:
ctx.close_position()
# Setups: each higher low (lower high) that a trend confirms, until it ends.
if sl is not None:
self.buy_setup = {"stop": sl["level"], "trigger": sl["trigger"]} if trend == 1 else None
if sh is not None:
self.sell_setup = {"stop": sh["level"], "trigger": sh["trigger"]} if trend == -1 else None
if trend != 1:
self.buy_setup = None
if trend != -1:
self.sell_setup = None
if self.buy_setup is not None and l <= self.buy_setup["stop"]:
self.buy_setup = None
if self.sell_setup is not None and h >= self.sell_setup["stop"]:
self.sell_setup = None
# The signal: the first close beyond the other end of the setup's swing candle. Each
# setup is used once; one that triggers while a trade is open is dropped.
buy = self.buy_setup is not None and c > self.buy_setup["trigger"]
sell = self.sell_setup is not None and c < self.sell_setup["trigger"]
setup = self.buy_setup if buy else self.sell_setup if sell else None
if buy:
self.buy_setup = None
if sell:
self.sell_setup = None
if ctx.position is not None:
return
if not ctx.gate("higher_low_or_lower_high_confirmed", setup is not None):
return
side = 1 if buy else -1
if self.view is not None:
if not ctx.gate("with_higher_chart_trend", self.trend(self.higher) == side):
return
stop = setup["stop"]
size = ctx.size_by_risk_pct(0.01, stop=stop, entry=c)
if not ctx.gate("size_at_least_the_minimum", size > 0):
return
tp = c + side * p.target_r * abs(c - stop) if p.target_r > 0 else None
self.exit_level = stop
if side > 0:
ctx.buy(size=size, sl=stop, tp=tp, tag="higher_low_buy")
else:
ctx.sell(size=size, sl=stop, tp=tp, tag="lower_high_sell")
def read_higher_chart(self, ctx, n, counts):
"""Feed each newly closed higher-chart candle into its own swing points."""
if counts:
self.counted = (self.counted + [ctx.bar_time])[-3:]
v = self.view
if v.bar_index < 0 or v.bar_index == self.view_seen:
return
self.view_seen = v.bar_index
end = v.bar_time # the higher-chart candle's close time
start = end - self.view_ms
if not any(start <= t < end for t in self.counted):
return # every chart candle in it was skipped
if v.high[0] > v.low[0]:
self.add_candle(self.higher, n, v.bar_index, v.high[0], v.low[0])
Your turn
Test the higher highs and higher lows strategy yourself
Change the settings, the market or the stop-loss and see the result in seconds. Ask the BacktestingMax AI: Backtest this on the DAX, 4-hour chart: swing points with 3 candles on each side. In an uptrend, buy after a candle closes above the high of the candle that made a new higher low. Stop-loss at the higher low, close the trade when a lower low forms, risk 1% per trade
.
Free account. No credit card needed.
See every strategy we tested Browse strategies shared by tradersBacktest on historical data with trading costs, run by BacktestingMax. Past results do not guarantee future results, and nothing here is financial advice. Updated 9 November 2026.