The trade history is the list of every closed trade on an account: the instrument, direction, open and close times and prices, volume, commission, swap and profit. It is the only complete record of a strategy's behaviour, and everything else, the return, the win rate, the drawdown, is calculated from it.
Getting it out
In MetaTrader 5, the History tab of the Toolbox lists closed trades. Right-clicking allows the period to be set and the list to be exported as a report or saved to a spreadsheet format. A verified record on a third-party service provides the same list, usually with the calculations already done. For any serious assessment, get the raw list into a spreadsheet.
What to calculate
The basics: number of trades, win rate, average winner, average loser, expectancy, and the total after commission and swap. Then the shape: the equity curve, trade by trade; the maximum drawdown and its duration; the longest losing run; and the distribution of results, since a record where one trade produced half the profit is fragile in a way the averages do not show.
Then the behaviour: average holding time, and its range; the spread of trades across markets, since a "multi-market" record that made all its money in gold is a gold record; the time of day and day of week trades opened, which shows whether the strategy trades news, rollover or thin sessions; and, for every trade, whether a stop was in place from entry, which the history shows through the stop-loss column.
A history of 320 trades shows a 44 per cent win rate, an average winner of £160 and an average loser of £82: a positive expectancy of about £24 a trade. The equity curve rises with two drawdowns of 11 and 15 per cent. So far, good. Then the detail: 212 of the 320 trades are in one pair, and 31 of them, all in a single month, opened within two minutes of high-impact releases and account for 40 per cent of the year's profit. The strategy is one pair, and a large part of its edge is news spikes on demo fills. On a live account, that month does not repeat.
Read an EA's trade history before its summary, and read it for behaviour, not just results. The questions are: does every trade have a stop; are the trades spread across the markets and periods claimed; is the profit spread across many trades or concentrated in a few; do the holding times match the description; and does the equity curve include losing periods of the kind that the market will certainly deliver again.
A history that answers all of those well is worth more than any return figure. A vendor who will not provide one has answered the question already.