How to read an MT5 backtest report

• Net profit alone says nothing; it only means something beside the drawdown it cost and the number of trades it took.
• Profit factor is gross profit over gross loss. One is break-even before costs.
• Recovery factor under one means the worst drawdown was larger than everything the run made.

A MetaTrader 5 Strategy Tester report is about forty numbers, and most people read three of them. This guide covers the ones that change a decision, using MetaQuotes’ own definitions, and says what a value has to be before it should stop you.

No figure from any XauStack strategy appears below. The point is to be able to judge a report — including ours — rather than to be told an answer.

Total net profit

MetaQuotes defines it as “the financial result of all trades” (MetaTrader 5 Help). It is the end of the balance line minus the deposit, and on its own it is the least informative number on the page.

Two runs can end at the same net profit having taken completely different risk, completely different numbers of trades, and completely different amounts of luck. Read it last, not first, and read it as a percentage of the deposit rather than as dollars — a report at one deposit and the same report at four times the deposit will show the same money and a very different experience.

Drawdown: balance, equity, absolute, maximal, relative

The report gives several. The two that matter are the maximal ones, and of those the equity version.

Balance drawdown maximal is “the difference in deposit currency between the highest local balance value and the next lowest account balance value”, and equity drawdown maximal is the same calculation applied to equity rather than balance (MetaTrader 5 Help). Balance only moves when a trade closes. Equity moves while it is open. A strategy that holds losing positions for a long time can therefore show a comfortable balance drawdown and an equity drawdown that would have closed the account — so the equity figure is the honest one, and it is the one to compare against your broker’s margin call level.

The percentage in brackets is relative to the peak equity at the time, not to the deposit. That matters when the run made money: a 30 % equity drawdown late in a profitable run is a much larger dollar figure than a 30 % drawdown in the first month.

A bad value is any equity drawdown you would not sit through. That is a personal number, not a universal one, but there is a hard edge: a maximal equity drawdown near or past 50 % means the run was, at some point, one bad week from over.

Profit factor

“Ratio of the gross profit to the gross loss. A value of one means that these parameters are equal” (MetaTrader 5 Help).

So:

• Below 1: the run lost money. Nothing else on the report can rescue that.
• Exactly 1: every dollar won was matched by a dollar lost, before the costs the tester did not charge. In practice that is a losing strategy.
• Just above 1: fragile. A profit factor of 1.05 means a five per cent shift in costs or fills takes the whole edge, and both spread and commission move by more than that in real trading.

Higher is better, but a very high profit factor on very few trades is usually a small sample rather than a strong edge. Read it with the trade count, always.

Recovery factor

“The amount of money risked by the Expert Advisor to make the profit it obtained… calculated as the ratio of gained profit to the maximum drawdown” (MetaTrader 5 Help).

This is the number that turns net profit into something comparable. A recovery factor of 1 means the strategy made, over the whole window, exactly as much as its worst single drawdown took away — you would have had to endure the entire gain as a loss at some point to collect it. Below 1 is worse than that.

It is also the number that punishes a good-looking equity curve with one catastrophic hole in it, which is precisely the shape a grid or averaging strategy tends to produce.

Total trades

Not a quality measure — a confidence measure. Twenty-four trades over eight months is not evidence of anything, however good the other numbers look; a single lucky sequence can produce all of it. Several hundred trades across several market conditions is a sample you can start to reason about.

When a report shows very few trades, treat every other figure on it as provisional and go and find more window.

Win rate, expected payoff, Sharpe

Win rate on its own is meaningless: a strategy can win nine times in ten and lose money on the tenth, and several families are designed to do exactly the opposite. Read it beside expected payoff, which MetaQuotes describes as “the average return of one deal” — that is the number that already combines how often it wins with how much.

The Sharpe ratio on a tester report is calculated on the run’s own returns and is comparable between two runs of the same length on the same data, and not much further than that.

A reading order that works

1. History Quality and the date range — is this evidence at all?
2. Total trades — is there enough of it?
3. Equity drawdown maximal — could you have survived it?
4. Profit factor and recovery factor — was the edge real and worth the hole?
5. Net profit — last.

If a report fails at step 1, 2 or 3, the rest of it is decoration.

Before you act on any of this

Every figure XauStack publishes comes from the MetaTrader 5 Strategy Tester, never from live trading, and none of it is a forecast. A backtest describes one data set over one window with modelled fills and, often, no commission charged. Read the risk disclosure, then generate a report of your own with how to backtest a gold EA in MT5.