Since we can't know which trades will be profitable, if the signals meet your trading system there is no reason to try and cherry pick. Doing so invalidates your strategy and makes your odds less than radom.
Trading correlated pairs can work but again this should be part of your strategy, if you take the time in evaluating pairs you will find that some correlated pairs move before the other correlated pair. This could be used as part of a strategy. However, many of these correlations have less volatility so the move is a fraction of the main correlated pair, as such use cautiously.
Assuming, your talking about trades that meet your strategy than there are 2 things to consider:
1) This is when using high leverage is beneficial by allowing you to take multiple trades and having the margin to cover them, Do Not leverage your positions, us the leverage to equalize your position sizing.
2) If you're trading a pair with a pip value of 1$ than size your position so all your trades have a value of or as close to 1$ as possible. This is important to insure a loss is made up with 'A' win, rather than a win and a half for example. Nothing worse than to have a 20 pip loss or 20$ loss and a 40 pip winner but the winner is worth .881 per pip so a profit of 35.24$ rather than 40$ or 39$ or 41$ depending how you position your size.
I think to many traders see leverage as a way to get rich rather than a way to go broke, however, using leverage in a beneficial manner makes a world of difference as well as reducing your overall drawdown by allowing you to equalize your position size.
Above all else follow your system this is the only way to truly evaluate your performance and that of your system.