Run a backtest and you get a wall of numbers: win rate, profit factor, expectancy, average win, average loss, max drawdown, Sharpe, total trades. Most beginners glance at the net profit and the equity curve and stop there — which is exactly how losing strategies get funded. The numbers that matter are not the flashiest ones.
This guide decodes a backtest report line by line, using real figures from our own published BTC and ETH research, so you can tell at a glance whether a strategy has an edge or just got lucky.
Total trades — the number that validates all the others
Before you read a single performance metric, check the sample size. A strategy with 8 trades tells you almost nothing; the stats are noise. A strategy with several hundred trades produces numbers you can trust. Always read total trades first — it is the confidence dial on everything else.
In our research, sample sizes ranged from a single trade (a strict ICT Silver Bullet setup that fired just once in a year) to over 15,000 (a 1-minute scalper). A one-trade result is a coin flip dressed up as data; never act on it.
Win rate — the most over-rated metric
Win rate is simply the percentage of trades that made money. It is the number beginners chase and the one that misleads most, because it says nothing about how big the wins and losses are. A 40% win rate can be wildly profitable; a 70% win rate can blow up an account.
In our study, an RSI mean-reversion system on ETH won 40.6% of trades — and still lost 87% over the year, because the losers were larger and costs ate the rest. High win rate, dead strategy.
Average win vs average loss — the payoff ratio
This pair tells you the other half of the story win rate ignores: how much you make when right versus lose when wrong. Expressed as a ratio (or in R-multiples, where 1R is your initial risk), it is what lets a low win rate still print money.
A strategy that wins 35% of the time but makes 3R per winner and loses 1R per loser has a strong positive expectancy. A strategy that wins 65% but makes 0.5R and loses 1R is a slow bleed.
Profit factor — the quickest health check
Profit factor is gross profit divided by gross loss across every trade. It answers, "for every dollar I lost, how many did I make?" Above 1.0 is profitable; 1.5 to 2.0 is generally solid; sustained values far above that over a big sample are rare and worth double-checking for overfitting.
Below 1.0 means the strategy loses money, full stop. Every one of the 60+ baselines in our research landed under 1.0 after costs — the very best, a previous-day-level break on ETH, reached only about 0.996. That is the cleanest one-glance tell that an idea has not yet earned its keep.
- Profit factor 1.0 → breakeven before you even count slippage.
- 1.5–2.0 → generally considered solid over a meaningful sample.
- Far above 2.0 → rare; suspect overfitting or too few trades.
- Below 1.0 → a losing strategy, no matter how nice the curve looks.
Expectancy — the bottom-line number
Expectancy is the average profit or loss you can expect per trade, combining win rate and payoff into one figure. Positive expectancy means the strategy makes money over a large enough sample; negative means it loses, period.
It is the single most decision-useful number in the report. In our data, an opening-range breakout on BTC showed an expectancy of about −$10.76 per trade across 417 trades — a small, consistent leak that, multiplied out, drained nearly half the account. Tiny negative expectancy still compounds into ruin.
Maximum drawdown — the survivability test
Max drawdown is the worst peak-to-trough fall in equity over the test. It answers a question profit cannot: could you actually have sat through this without quitting or blowing your risk limits? A strategy that nets a profit but drew down 60% on the way is one most humans abandon at the bottom.
Several systems in our research hit drawdowns near 100% — total account destruction. A profitable-looking average return is meaningless if the path there wipes you out first.
Sharpe and the equity curve — context, not gospel
Sharpe ratio gauges return per unit of volatility; negative Sharpe (as nearly every strategy in our study showed) signals returns that did not justify the risk. The equity curve is a visual — useful for spotting whether profits came from one lucky streak or steady edge, but easy to fall in love with.
Read the curve to ask "is this smooth and broad, or one spike over too few trades?" Then go back to the hard metrics. A pretty curve over 20 trades is decoration; a positive expectancy over 400 trades is evidence.
Frequently asked questions
Which backtest metric matters most?
Expectancy, read alongside total trades. Expectancy is the average profit or loss per trade and tells you whether the strategy makes money at all; the trade count tells you whether to trust the number. Win rate alone is the most over-rated metric.
What is a good profit factor in a backtest?
Above 1.0 is profitable; roughly 1.5 to 2.0 is generally considered solid over a meaningful sample. Below 1.0 is a losing strategy. Values far above 2.0 are rare and often signal overfitting or too few trades.
Why does a high win rate not mean a good strategy?
Because win rate ignores trade size. A 70% win rate with tiny wins and a few huge losses still loses money, while a 35% win rate with large winners can be very profitable. Always pair win rate with the payoff ratio and expectancy.
What does maximum drawdown tell me?
It is the worst peak-to-trough drop in equity, which measures whether you could realistically survive the strategy. A profitable backtest with a 60% drawdown is one most traders abandon at the bottom, so drawdown is as important as return.