Trendline Trading Strategy Backtest: Do Trendline Breakouts Work? (Dow Futures)

00The verdict in three lines

• Do trendline breakouts work? Barely, and not reliably: on Dow futures 1-hour bars the breakout made +$49 a trade after costs, at the 95.0th percentile of random entries, but +$59,714 of its +$63,591 came from 2020, and it lost money without its 10 best trades.
• Do trendline bounces work? No: they lost before costs (-$10 and -$22 a trade), and only 12% and 10% of bounce trades won.
• Buried, all four: Fragile (the breakouts), No proven edge (the bounces).

01The trendline strategy in the video

The source is a YouTube masterclass, “How To Trade TRENDLINES In UNDER 17 Minutes” (Tori Trades, about 1.9 million views), which marks up micro Dow futures (MYM) from the monthly chart down to the 1-hour chart and trades there. Its rules, in its own terms:

• Upward trendline: point A is the lowest point on the chart; point B is the most recent higher low that leaves the line unbroken. A steeper line then starts at the old point B and goes to the next unbroken low, so the lines fan out. Downward trendlines mirror this on the highs.
• Bounce: when price comes back to a line, trade in its direction, with the stop on the other side of the line.
• Breakout (the presenter's favourite): when price breaks a line, trade the break, with the stop on the other side of the opposite line.
• Exit: no take-profit; trail the stop along the line and get out when price breaks it.

The video's only real trade is a single example on crude oil. The presenter draws the lines by eye; we test the rules as stated, made mechanical, and the article says where our reading had to choose.

02How we drew the trendlines (the mechanical reading)

• Swing lows and highs: an hourly bar lower (higher) than the 5 bars, or in the second version the 10 bars, on each side; it is only known once those later bars have closed.
• The chart: the last 2,000 hourly bars (about 17 weeks). The monthly-to-hourly markup is replaced by this long window: the big highs and lows are also hourly swing points, and the fan always ends on the steepest line near price.
• Unbroken: no hourly close beyond the line from point A to now. The active line is the steepest one of the fan; lines are measured in bars, so weekends do not bend them.
• Breakout: an hourly close through the active line. Bounce: a bar that reaches the line and closes back on its side. Entry: the next minute's open. Stop: one tick beyond the guiding line (the opposite line for a breakout, the touched line for a bounce), moved along it every hour and never loosened. No target, no time limit.

With 5-bar swings an upward line existed on 87.5% of hourly bars, but the active line changed often: a median of 8 hours before it was broken or replaced by a steeper one. Trendlines drawn this way are redrawn all the time, as the video itself says.

03How we tested it

YM (E-mini Dow futures, $5 a point), NinjaTrader 1-minute history, back-adjusted, January 2020 to 25 September 2026: 39,623 hourly bars. This is the first test on YM on this site. Costs: $15.76 a round trip per contract (commission plus one tick of slippage on each fill), plus one more round trip for each quarterly roll held. Periods: 2020 to 2023, 2024 to March 2025, and a holdout from April 2025 whose months were already used by our ES and NQ tests (the YM file itself is new). First-period bar t ≥ 2.5. Control: every entry moved to the same hour of another session, with the same stop-and-trail rule on the lines of that time, 500 times. Engine check: swing points, the fan on hand-made data, signals, the trailing stop, an engine shortcut checked against a full redraw at every bar, and an independent re-implementation that redrew every line at every bar and matched all 5,050 trades (tl_selftest_result.txt).

04Results: does trendline trading work?

Cumulative net dollars per YM contract by month. * Holdout months already used on ES and NQ.

The risk behind each trade

The breakout's stop, on the other side of the opposite line, is far away: the video's example risked $50, our median breakout risked $147 on a micro contract. The bounce's stop, one tick beyond the touched line, is close, and it was hit within the hour most of the time. The breakout made money on longs (+$63,837) and about nothing on shorts (-$245).

05Trendline signal or random entry?

The 5-bar breakout reached the 95th percentile exactly, the only version to pass this gate; the others were in the middle of the random results.

06By period and by year

* The holdout months were already used by our ES and NQ tests; this is the first test on the YM file, so the holdout is not clean.

The 5-bar breakout made +$59,714 in 2020, the crash and the recovery, and about nothing over the following years together; it was positive in the holdout (+$10,521). The 10-bar breakout lost in the holdout.

07The gates

The data-gap check removed 13, 10, 4 and 3 trades open across a known hole in the data (the largest, 5 to 16 April 2023, is missing from NinjaTrader's server); no verdict changed. Fragile: positive, but negative without its 10 best trades. No proven edge: negative before costs.

TWhat this shows, and what to take into your own tests

What the results show

• Written down as the video describes them, trendline breakouts on Dow futures 1-hour bars made +$49 a trade after costs with 5-bar swing points, and only just beat the same trades at random times (95.0th percentile). +$59,714 of the +$63,591 came from 2020.
• Trendline bounces lost before costs: the stop one tick beyond the touched line was hit within a median of 38 minutes, and 12.3% of bounce trades won.
• The breakout's stop, on the other side of the opposite line, sat a median of 294 points away: $1,471 on one YM contract, $147 on one micro (MYM), far more than the $50 of the video's example.

What remains unknown

• The trader's own eye. She draws and redraws the lines by hand; another drawing could give other trades, and that cannot be tested by anyone.
• The optional take-profit at support or resistance, which the video allows; we tested the version she prefers, letting the trend run.
• Other markets (the video's live example is crude oil) and other timeframes.

Why it was buried

The breakout's result rests on one year: without 2020 it is about flat, and it was negative without its 10 best trades (-$41,790). Its first-period t-statistic was 1.27 against the 2.5 we require, and the second period lost money. The bounce lost money before any cost.

What you can apply to your own tests

• Write your lines down as rules before you trade them: where point A is, which low is point B, what counts as a break. If you cannot, you cannot test them.
• Measure the stop in dollars before the trade. A stop on the other side of a line can be hundreds of points away.
• Check the result year by year. One exceptional year can carry a whole backtest.

A note on the numbers: drawdown figures are measured on closed trades unless a table says it includes open losses. Open losses inside a trade are not counted in a closed-trade drawdown, so real account risk is higher, especially for systems with no stop. Backtests are not live trading.

FAQCommon questions

Does trendline trading work?

Not clearly, in our test of the rules a popular YouTube masterclass states. The breakout version only just beat random entries and depended on 2020; the bounce version lost before costs. The presenter draws lines by eye, which no backtest can reproduce.

How do you draw a trendline correctly?

The method we tested: an upward line from the lowest point to the most recent higher low that leaves it unbroken, then steeper lines from each new point; downward lines the same on highs. Written as rules, the active line changed every 8 hours or so on 1-hour bars.

Is a trendline breakout or a bounce better?

In our test the breakout did better: +$49 a trade after costs against -$26 for the bounce, whose stop one tick beyond the line was hit within a median of 38 minutes. Neither passed our gates.

Where should the stop go when trading trendlines?

The video puts it on the other side of the line. For breakouts, on the other side of the opposite line, that was a median of 294 points: $1,471 on one Dow future (YM) or $147 on a micro (MYM).

How was the trendline strategy tested?

On YM (Dow futures) 1-minute data from 2020 to 2026, built into 1-hour bars, one contract, with rules written and dated before the data was run, costs included on every trade, results split into development, validation and holdout periods, a random-entry control as the benchmark, and every variant reported. Read the protocol.

08Check our arithmetic

Code provenance: our own code, written from the specification.

09Data notes

Source. NinjaTrader YM 1-minute history (YM DEC26, merge back adjusted), exported on 22 October 2026 and checked before use: no duplicate or out-of-order bars, no price errors, and contract rolls adjusted (the session-open gaps match the back-adjusted ES file). Holes are listed in data_holes_ym.json and never filled. Fills: the next minute's open for entries; the stop at its level, or a minute's open if it gapped through; one tick of slippage per fill is in the cost.

10Corrections

None since publication.

+More tests and guides

• Nr7 trading strategy
• Donchian channel breakout strategy backtest
• Do stop losses work
• Trend following vs mean reversion
• Free prop firm challenge calculator
• What all the backtests found so far
• The backtesting protocol
• How to backtest a trading strategy without fooling yourself
• Backtesting glossary