Compare every backtest with simply holding
A backtest that made 18% sounds good until you learn the market rose 30% over the same months. The comparison with holding is the first check, not the last.
A result needs something to stand next to
A backtest ends with a number: the rule turned $1,000 into $1,180, so it made 18%. On its own that number tells you very little. You do not yet know whether the rule added anything, or whether the market did all the work while the rule just happened to be in it.
The simplest thing to stand next to it is "buy and hold": buy on the first day of the test, do nothing, and sell on the last day. It needs no skill and no screen time. If a rule cannot be told apart from that, the rule has not shown anything yet.
A worked example
Say the test covers one year and the market went from 100 to 130 in that year. Holding $1,000 through it ends at $1,300, a gain of 30%, minus one entry fee and one exit fee.
The rule ended at $1,180. It made money, and it still finished $120 behind doing nothing. Anyone who looked only at the 18% would have missed that.
Now turn it round. In another year the market falls from 100 to 80. Holding ends at $800. A rule that ends at $950 lost 5% while the market lost 20%. It lost less than holding, and it still lost.
What time in the market changes
Holding is exposed to the market every single day. Most rules are not. Suppose the rule above held a position on 40% of the days and sat in cash on the other 60%.
That changes how to read the 18%. The rule earned it while carrying risk for less than half the time. For the other days the money was not exposed to a sudden fall. Whether that matters to you depends on what you want: the largest final number, or a smaller number with less time at risk.
It also cuts the other way. A rule that is in the market 95% of the time and matches holding has done almost nothing except pay more fees.
Compare the falls, not only the gains
Put the largest drawdown of each side by side. Drawdown is the biggest fall from a peak to a later low. If holding fell 25% at its worst and the rule fell 8%, the two were very different to live through, even if the final numbers were close.
A plain way to write the result is in three lines: the return of the rule and of holding, the worst drawdown of each, and the share of time the rule was in the market. Three lines say more than one headline number.
Make the comparison fair
Use the same first and last day for both. Use the same starting money. Charge the same fees and slippage on the rule's trades as you would pay in real life, and charge holding its one entry and one exit.
Then repeat the comparison on other periods and other markets. A rule that beats holding in a falling year and trails it in a rising year may simply be a rule that is often in cash. That is a property worth knowing about, and it is not the same thing as skill.
What the comparison does not prove
Beating holding in one test does not mean the rule will beat it next year. One period is one sample, and a strong trend in either direction can flatter or punish any rule.
Trailing holding does not make a rule useless either, if what you wanted from it was smaller falls. The comparison does not hand you a verdict. It stops you from reading a rising market as your own good idea.
The Candlesview backtester shows the buy-and-hold result for the same period next to the strategy, so the two can be read together.
To take away
- A backtest return means little until it sits next to holding over the same dates.
- Read return, worst drawdown and time in the market together.
- Beating holding in one period is one sample, not proof.
Try it
Keep reading
A scan matched 8 times and price rose after 6 of them. That feels like 75%. Here is why it says almost nothing, and what to compare it with before believing it.
20 November 2026Testing ideas · 5 minMixing timeframes without looking into the futureA daily trend filter on an hourly chart is a sound idea that is easy to test wrongly. The rule is simple: a higher-timeframe candle only counts once it has closed.
19 November 2026Order flow and derivatives · 5 minWhat a liquidation map can and cannot tell youThe bright bands on a liquidation map are estimates built on assumptions about leverage. They show where pressure may sit, not where price is going.
18 November 2026Education only. Nothing here is financial advice or a recommendation to buy or sell anything; examples are arithmetic, not results. Trading involves a high risk of losing money.