Beat Buy AndNovember 6, 202615 min read

How to Beat Buy and Hold: A Fair Test for Any Strategy

Beating buy and hold means more than a higher return. Learn how to pick a fair benchmark, measure risk and costs, and avoid overfitting before you trade.

A strategy beats buy and hold when, over the same market and period, it earns a better return for the risk you took, after costs and taxes, and when that edge holds up on data it was never tuned on. A higher headline return alone does not count. Most strategies that look better than holding an index fail one of those tests, and this guide shows you how to run them.

TL;DR

Buy and hold is the default benchmark: buy an asset or index, keep it through the ups and downs, and ignore short-term signals.
Pick the benchmark before you run the test, and match it to what your strategy holds and the risk it takes.
Judge the result on return, maximum drawdown, Sharpe ratio and costs together, not on total profit.
Overfitting is the most common reason a backtest beats buy and hold on paper and fails in live trading.
A backtest is evidence about the past. It cannot guarantee the next price move.

What "beating buy and hold" actually means
Choose a fair buy-and-hold benchmark first
The metrics that show whether you really won
Why active strategies win in some markets and lose in others
Backtest against buy and hold without fooling yourself
Test popular strategies on real history with Backtestify
Can a backtest predict the next move?
A quick pre-trade scorecard
FAQ
Sources

What "beating buy and hold" actually means

Buy and hold is a passive approach: an investor buys stocks or ETFs and keeps them for a long period, without reacting to short-term price moves or technical indicators. It costs almost nothing to run, needs no timing decisions, and captures whatever the market delivers.

That is why it is the bar. If you trade actively, you take on extra work, extra transaction costs and extra ways to be wrong. The only reason to do it is that the result is better than what you would have gotten by doing nothing. "Better" has three parts:

Higher return over the same period and the same market.
Equal or lower risk, usually measured by maximum drawdown and volatility.
Still better after costs: commissions, spreads, slippage and taxes.

A strategy can win on one and lose on the others. A trend filter that sits in cash during crashes might earn less than holding but cut the worst loss by a lot. Whether that is a win depends on what you need. The point is to decide the yardstick in advance rather than choose it after you see the numbers.

Choose a fair buy-and-hold benchmark first

The most common error in these comparisons is an unfair benchmark. Comparing a small-cap momentum strategy to a bond index, or a leveraged futures system to an unlevered stock fund, tells you nothing.

A good rule: the benchmark should share your strategy's investable universe and roughly its market exposure and risk. In practice:

Your strategy	A fair buy-and-hold benchmark
Long-only US large-cap stocks	S&P 500 total return
Trades one stock or one ETF	Holding that same asset over the same dates
Sector rotation or a sector bet	The matching sector index
Market-neutral long/short	The risk-free rate (such as T-bills)
Mixed stock and bond allocation	A blended portfolio such as 60/40

This mapping follows the logic in benchmark selection guidance for backtests, and the professional standard is similar. The CFA Institute's GIPS standards say a valid benchmark is specified in advance, relevant, measurable, unambiguous, investable and complete.

Use total return, not price only

If your strategy's results include dividends, the benchmark must too. The GIPS guidance states that comparing a total-return strategy with a price-only benchmark is misleading, so benchmark returns should be total returns. A price-only index understates buy and hold by roughly the dividend yield every year, which flatters any strategy compared against it.

Do not rely on a single start date

For a single-asset test, the comparison is usually holding that asset. But that baseline assumes you bought exactly on the first day of the test, which is rarely realistic. One practitioner suggests adding a second benchmark, such as a 60/40 or 50/50 equity-bond portfolio, to give a more realistic baseline. Testing several start dates is a simple extra check: if the strategy only wins when the test begins right before a crash, the edge is the start date.

Lock the benchmark before you see results

Write down the benchmark, the dates and the metrics before you run anything. Institutional guidance says a benchmark should be constructed before the evaluation period so the result is not judged against something built after the fact. Retail traders have the same temptation: if the first benchmark looks bad, it is easy to swap in a weaker one. Do not.

The metrics that show whether you really won

Total profit is the least useful number on the report. Look at these together.

Five metrics for comparing strategy performance

Annualized return (CAGR). Convert both results to a yearly rate so a 12-year test and a 3-year test are comparable.

Maximum drawdown. The largest peak-to-trough loss. Buy and hold on a stock index can fall by half or more in a bear market. A strategy that earns slightly less but never falls as far may be the better fit for your nerves and your account.

Sharpe ratio. Return per unit of volatility. It lets you compare strategies with different risk levels. An academic paper from the University of Ulm notes that on a Sharpe-ratio basis, buy and hold behaves like other fixed-weight allocations, which makes it a reasonable benchmark and not an arbitrary one.

Profit factor and win rate. Gross profit divided by gross loss, and the share of winning trades. These describe the strategy's shape but say nothing about the benchmark, so never use them alone to claim you beat holding.

Costs. Include commissions, spreads and slippage. A study of moving-average strategies on ETFs found that, relative to buy and hold, they had lower average returns and Sharpe ratios, and became less profitable when implemented with ETFs rather than underlying indices, although they fared better on factor-adjusted measures such as CAPM alpha. That is a good illustration of why one metric can mislead: the same strategy looks different depending on which number you read and how you implement it.

The same Ulm paper makes a point worth remembering: a rational investor should choose buy and hold unless the data show sufficiently strong evidence that the alternative's risk-adjusted performance is better. The burden of proof sits with the active strategy.

Why active strategies win in some markets and lose in others

No approach wins everywhere. Work by Perold and Sharpe, discussed in an Investing.com analysis, shows that bull markets favor buy and hold, sideways markets favor constant-mix strategies, and bear markets favor portfolio-insurance approaches. A strategy's relative performance also depends on how crowded it is.

That has a practical consequence. If you test only on a long bull run, almost everything that stays long beats a defensive system, and holding wins by default. If you test only on a choppy range, a mean-reversion system looks brilliant. Your test window should include:

A strong uptrend.
A drawdown of 20% or more.
A sideways stretch.
A volatility spike.

The research on what can beat holding is mixed, and honest. A large test of more than 4,100 strategies on the S&P 500 found that only technical strategies exploiting underreaction and overreaction dominated buy and hold in some simulation setups. A long-run study of trend following on the Dow Jones Industrial Average and international indexes found excess returns that were substantial but fading, non-persistent and highly sensitive to methodology. Takeaway: edges exist in specific settings, they can decay, and the way you test changes the answer.

Behavioral patterns are one reason some rules can work at all. Traders who sell winners too early and hold losers too long are showing what researchers call the disposition effect, and a rule-based strategy takes that habit out of the picture. Rules do not make a strategy profitable, but they make it testable.

Backtest against buy and hold without fooling yourself

The fastest way to "beat" buy and hold is to overfit. Try enough rule variations on the same history and one will look great by chance. Bailey and López de Prado show that backtest overfitting occurs when too many variations are tried relative to the data available, and that as few as three independent trials can produce a likely-false strategy. They propose the deflated Sharpe ratio, which adjusts for sample length, skewness, kurtosis and the number of variations explored. For a plain-language look at the same trap, see this explainer on overfitting in trading.

A checklist you can follow:

Write the rules in full before testing. Entry, exit, position size, costs. No changes mid-test without logging them.
Fix the benchmark and the metrics up front. See above.
Count your trials. Every parameter tweak and every indicator swap is an attempt. The more you try, the higher the bar the final result must clear.
Split the data. Develop on one period and confirm on a later period you have not touched. Our guide to walk-forward analysis covers this in detail.
Keep parameters few. Fewer knobs means less room to fit noise. Our piece on strategy optimization without overfitting shows how to improve rules without memorizing history.
Use realistic costs and fills. A scalping or high-turnover idea can lose its entire edge to spreads.
Compare on the same data. Same dates, same instrument, same price type (adjusted for splits and dividends).
Check the number of trades. A handful of trades is an anecdote, not a sample.

If you want the full process from rule to report, read how to backtest a trading strategy step by step and our deeper look at spotting and stopping overfitting.

A worked example of the thinking

Suppose your rule-based system on an index ETF returns 7% a year with a maximum drawdown of 18%, while holding the ETF returned 9% with a maximum drawdown of 34% over the same dates. Did you beat buy and hold?

On raw return, no: you gave up two points a year.
On risk, yes: you took roughly half the worst loss.
On risk-adjusted measures, it depends on volatility and the Sharpe ratio, which you should calculate.
After costs and taxes, the gap in raw return may widen, because active trading generates more taxable events than holding.

None of these answers is wrong. What would be wrong is announcing a win because the drawdown number is nicer, or hiding the lower return. State the trade-off plainly and let the goal decide: maximum growth, or a smoother ride you can actually stick with.

This is also the right lens for strategies you see online. A popular setup from a video may show impressive trades, but unless someone has run it over a long window against the matching benchmark, with costs, you have a story and not a result.

Test popular strategies on real history with Backtestify

Backtestify is built for exactly this check. You describe a strategy in plain English, run it on real historical market data, and get a report with win rate, profit factor, maximum drawdown and a trade-by-trade breakdown, so you can see how it behaved across real market conditions. You can then improve the rules and compare the original and improved versions side by side on recent data, and export the result as TradingView Pine Script.

We also publish results for both winning and losing strategies, including setups popularized by well-known trading creators. Browse the published strategy library to see how reports look, for example the Tori Trades trendline strategy or the sweep, market-structure shift and fair value gap setup.

There is a free plan to start with. Pro unlocks unlimited backtesting, improvement and forecasting at $29 per month or $190 per year. You can see the plans on our homepage or start on the free plan.

Backtestify

Can a backtest predict the next move?

No. A backtest shows how a precisely defined rule would have performed in the past. It does not prove the future will resemble that history, and it cannot guarantee that a strategy will keep beating buy and hold. The research above points the same way: excess returns from active rules are often fading and sensitive to method, and a result that looks strong can be the product of many trials.

What a good backtest can do is narrow the field. It filters out ideas that never worked, shows how deep the losing stretches were, and tells you what to expect so a drawdown does not surprise you. Forecasting tools that report probabilities and confidence levels are most useful when treated the same way: as a measured read on odds, not a promise. Before you put real money behind any rule, run it on data it has not seen, size positions so a bad run is survivable, and re-test as conditions change.

If your plan involves swing trades or short-term setups, our guides to swing trading backtests and price action backtesting walk through how to check for a false edge in those styles. For a broader view of rigorous methodology, see our backtesting blog.

A quick pre-trade scorecard

Before you trust a strategy that appears to beat holding, answer these:

Question	What you want to see
Was the benchmark chosen before the test?	Yes, and documented
Does it match the strategy's universe and risk?	Yes
Are returns total return on both sides?	Yes
Are costs and slippage included?	Yes, realistic
Does it win on risk-adjusted terms, not just profit?	Yes
Does it hold on out-of-sample data?	Yes
Does it work across more than one market regime?	Yes
Is the trade count large enough to mean something?	Yes

If you can tick every row, you have earned the right to take the strategy seriously. If you cannot, you have found out cheaply, which is the whole point of testing first.

FAQ

Can an active strategy consistently beat buy and hold?

Some rules beat it in specific periods or markets, but the evidence shows those edges are often fading and depend heavily on how the test is built. Treat any claim of consistent outperformance as a hypothesis to test on unseen data, with costs, against a matched benchmark.

What is the best benchmark for a stock strategy?

Use the one that matches what you hold. For a long-only US large-cap strategy, that is a total-return version of the S&P 500. For a single stock, hold that same stock over the same dates. For a market-neutral strategy, use the risk-free rate.

Is a higher return enough to say I beat buy and hold?

No. Compare maximum drawdown, Sharpe ratio and costs too. A strategy with a higher return but a much deeper drawdown, or one that only wins before fees, has not clearly beaten holding.

How do I know my backtest is not overfit?

Fix the rules and benchmark in advance, count how many variations you tried, keep parameters few, and confirm the result on a later period you did not use to build the rules. If the edge disappears out of sample, it was probably noise.