How do you know when your sample of trades is large enough?

A handful of profitable trades may be encouraging, but it may not provide enough evidence to establish a strategy's reliability. The sample required can depend on the strategy's frequency and variability. What would convince you that you've collected enough data to judge a new trading method?