Reading Your Results

The Overview tab, strategy vs buy-and-hold comparison, key stats, equity curve, and drawdown chart explained.

Reading Your Results

When a backtest finishes, the page scrolls down to the results section, which opens on the Overview tab. Five more tabs sit beside it: Charts, Return Breakdown, Trade Details, Performance by Condition, and Monte Carlo. This article walks through what each one shows and -- more importantly -- how to interpret it without fooling yourself.

Start With Three Questions

Before reading any individual number, get oriented with three questions:

1. Did the rules beat doing nothing?

The headline banner shows Total Return with a small line underneath comparing it to buy & hold -- what you would have made by simply buying the asset on day one and holding to the end. A strategy can be profitable and still lose to the do-nothing alternative. See Benchmarks & Relative Performance.

2. What did the ride cost?

Max Drawdown is the worst peak-to-valley fall along the way. Two strategies with identical returns can feel completely different to hold -- this number is the difference.

3. How many trades is this built on?

The Trades stat. Ten trades can look brilliant by pure luck; the more trades behind a result, the harder it is for luck to explain it. See Statistical Significance & Trade Count.

The Overview Tab

The Overview tab is a stack of cards, top to bottom. Every stat card has a small info icon -- hover it for a one-line explanation, and click through to the metric's Learn article for the full story.

Strategy vs Buy & Hold

Below the headline banner, the Strategy vs Buy & Hold table puts three rows side by side -- Total Return, Annualized Return, and Max Drawdown -- for your strategy and for the buy-and-hold benchmark, with a Delta column showing the gap. Two reading notes: annualized return converts the total into a per-year growth rate so different-length tests are comparable (see Total & Annualized Return), and a strategy can win on return while losing on drawdown -- or the reverse, which is often the more interesting outcome.

If you selected more than one leverage multiplier, a Leverage row of pills sits above the banner -- every card on the page switches to the multiplier you pick.

Transaction Cost Analysis

When your run includes commission, spread, or slippage, this card splits Gross Return (before costs) from Net Return (after) and itemizes the Total Cost Drag per cost type. If the card is absent, you ran a frictionless simulation -- treat the results as a ceiling, not an estimate.

The stat grids

Three grids of small cards follow. In brief, grid by grid:

Risk-Adjusted Performance

Sharpe Ratio (return per unit of up-and-down movement), Sortino Ratio (same idea, counting only downside moves), Calmar Ratio (annual return against worst drawdown), Alpha (what is left after accounting for what the asset itself did), and Beta (how strongly your equity moves with the asset).

Risk Metrics

Volatility (how bumpy the ride was), VaR (5%) (the loss level your worst 5% of candles reached), and CVaR (5%) (the average loss inside that worst 5%).

Trade Statistics

Win Rate, Trades, Profit Factor (gross wins divided by gross losses), Expectancy (average result per trade), Exposure (the share of time you actually held a position), Avg Win, Avg Loss, Risk/Reward, and Avg Hold Time.

Why is a stat showing N/A?

Some statistics are undefined rather than zero in certain runs. A backtest with no losing trades has no loss to divide by, a zero-trade run has no per-trade stats at all, and a fully wiped-out account has no meaningful annualized return. N/A means "this cannot be computed for this run" -- not "something went wrong".

The Charts Tab

The Charts tab holds two interactive charts. Both pan (drag) and zoom (scroll), and both show a crosshair readout of every visible series as you hover.

The price chart

The main chart draws the asset's candles against the right axis and your strategy's equity curve -- labeled Backtest (1x), one line per leverage -- against the left Backtest Results axis. The equity curve starts at 1.0 and tracks the growth of your starting capital as a multiple: 1.25 means +25%, 0.80 means -20%. Any indicators your conditions used are drawn too -- price-based ones (like moving averages) on the price axis, oscillators (like RSI) in their own panels below.

A dashed gray Buy & Hold line shares the left axis and also starts at 1.0: it is what your starting capital would be worth had you bought the asset on the first candle and held it to the last. It always holds the asset long, whichever direction your strategy trades, and it ends on the same Buy & Hold return as the comparison table on the Overview tab. Because the two lines share one scale, the gap between them is the strategy's lead or lag against simply holding. The candles are measured on the right axis instead, so how high the price sits against the equity curve on screen says nothing about which one did better.

Trades appear as markers on the candles: an upward triangle where the strategy bought, a downward triangle where it sold (for a long strategy, entry and exit respectively; for a short strategy, the reverse). Click any marker to open its trade card -- entry and exit times and prices, the profit or loss, and which rule triggered each end of the trade. When you are zoomed far out, crowded markers shrink to dots or hide entirely; zoom in and they return.

The drawdown chart

The second chart plots drawdown over time: how far below its own previous peak each series was at every moment. Zero means at a new high; -20% means one fifth below the best point so far. It shows the asset's own drawdown alongside one series per leverage for your strategy, which makes it easy to see whether your worst stretch coincided with the market's -- or was entirely self-inflicted.

How Charts Stay Smooth on Big Backtests

A backtest over minute candles can span hundreds of thousands of candles -- far more than a chart can draw responsively. When a result covers more than roughly 25,000 candles, Simula groups neighboring candles into single displayed candles before sending chart data to your browser. Each grouped candle keeps the open of its first real candle, the close of its last, its highest high, and its lowest low -- so the shape of the market, including every extreme, is preserved. All chart series (price, equity curve, drawdown) are grouped on the same grid so they stay perfectly aligned.

This is display-only. Every number everywhere else in your results -- total return, drawdown, Sharpe, the trade list, every statistic on every tab -- is computed from the full-resolution data before any grouping happens. The chart is a picture of your backtest; the numbers are the backtest.

Six real candles collapse into one displayed candle. The grouped candle keeps the true open, close, highest high, and lowest low -- but Trade 1's exit (second real candle) and Trade 2's entry (fifth real candle) now sit on the same displayed candle.

Grouping has one visual side effect worth knowing about. Trade markers attach to the displayed candle nearest their real timestamp -- so one trade's exit and the next trade's entry can land on the same displayed candle, even though they happened on different candles, possibly hours apart. It can look like the strategy closed and reopened within a single candle when it did not.

To see what really happened, click the markers: the trade card shows exact entry and exit timestamps, and the Trade Details table lists them to the second. Zooming in sharpens the chart's shape, but on a grouped result the finest unit the chart can show remains the grouped candle -- the trade card and trade table are the ground truth. (Results at or below the threshold are delivered at full resolution, with nothing grouped.)

Can two markers on one candle ever be real?

Yes, in one specific case: a stop-loss or take-profit can close a trade on the same candle that opened it, because stops trigger intra-candle the moment price touches the threshold. That single trade's entry and exit markers legitimately share a candle. The tell is the trade card: one trade with identical entry and exit candles is a genuine same-candle round trip; two different trade numbers are display grouping. See How the Simulation Works.

The Trade Details Tab

This tab opens with a Distribution of Trade Results histogram -- your trades bucketed by outcome, with a banner splitting them into losers, break-even, and winners. The shape matters as much as the win rate: many small wins with a few huge losses is a very different strategy from the reverse, even at the same total return.

Below it, the trade table lists every trade: Trade ID, Long/Short, Entry Date, Entry Price, Exit Date, Exit Price, Exit Reason (which rule or safety net closed the trade), and Profit/Loss. Columns sort, and Export CSV downloads the lot for your own analysis. The trade table and its CSV export come with the Standard and Premium plans.

The final trade and "Backtest End"

If your strategy was still holding a position when the data ran out, the engine closes it at the final candle's price so it can be scored, and labels that exit Backtest End. You will see the label in the trade table's Exit Reason column (always the last row, exiting on the final candle of your date range), on the trade's marker card on the chart, and in the Performance by Condition tab's exit column.

A Backtest End exit is not something your rules did -- it marks where the data stopped. The trade still counts in every statistic, and that is deliberate: if it sits on a large loss, that was real risk your strategy was carrying at the cutoff. The How the Simulation Works article explains the mechanics. Other exit reasons you will see in the same places: Stop Loss, Take Profit, Trailing Stop, Break-Even Stop, Time Exit, or the text of the exit condition that fired.

The Other Tabs

Return Breakdown

Period-by-period tables -- yearly, monthly, or daily -- comparing your strategy's return, trade count, and win rate against the asset in each period. Only populated if you switched the Breakdowns toggles on before running; the tables and their Export CSV come with the Standard and Premium plans. See Return Breakdowns.

Performance by Condition

Which of your rules actually pull the weight: how often each entry and exit condition triggered, and how every entry/exit combination performed (trades, win rate, average and total return). If one condition never fires, it is dead weight -- and this is the tab that tells you.

Monte Carlo

If you enabled Monte Carlo before running, this tab shows the range of outcomes produced by 1,000 simulated replays of your trades -- including the probability your account would have been ruined and whether your edge over buy & hold is statistically distinguishable from luck. Start with How It Works and Interpreting Results.

Some of these tabs are included on higher plans; locked tabs show a preview and say which plan unlocks them. A Share Your Results card below the tabs generates a public link to this result if you want to show it to someone. Anyone with the link can open it without an account and sees the Overview and Charts tabs. The chart shows which indicators your strategy uses, but not its entry and exit rules: the Return Breakdown, Trade Details, Performance by Condition, and Monte Carlo tabs stay locked on a shared link, and your conditions are not included in it.

Why does it say "No trades were executed"?

Your entry conditions were never all true on any candle in the range. Common causes: a threshold that price never reached (RSI below 20 on a calm asset), a crossover that never happened in your window, AND-ed conditions that never lined up on the same candle, or a date range that is simply too short. Loosen one value, widen the dates, or check that the comparison direction is what you meant -- then re-run.

Before You Trust a Good Result

A strong-looking backtest earns a skeptical second pass. Four checks, in order:

Enough trades?

Below a few dozen trades, luck can explain almost anything. A longer date range or a shorter timeframe gives your rules more chances to prove themselves.

Real costs?

Re-run with realistic commission, spread, and slippage and watch the Transaction Cost Analysis card. High-frequency strategies routinely go from impressive to unprofitable on this step alone.

Survivable drawdown?

Look at the drawdown chart and honestly ask whether you would have kept following the rules at the bottom of the worst valley. A strategy you would abandon mid-drawdown is a strategy you do not have.

Skill or luck?

Run it again with Monte Carlo enabled and check Is My Outperformance Real? And if you arrived at these settings by tweaking until the numbers looked good, read Overfitting & Curve Fitting before going further -- walk-forward analysis exists precisely to catch that trap.

Where to Go Next

The Metrics section -- every statistic on the Overview tab, each with worked examples and honest caveats.

Limitations & Pitfalls -- the ways backtest results mislead, and how to catch each one.

How the Simulation Works -- where these numbers come from, candle by candle.

Your First Backtest -- if you landed here without running one yet.