I love the thinking behind this question. Which is that there must be times when you should not take a trade because the probability of a successful outcome is already damaged before you even start.
Start with your parameters for identifying the simplest pattern - a trend. Once you have a panel of features you look for to firstly confirm a trend and secondly gauge it against other trends, then you can select the best trend to trade, either as a trend-following trade or a reversal trade. Note down what you would need to see in an ideal perfect uptrend, but keep it simple, just using a couple of time frames and a couple of MA’s. Make this into a check-list or score-sheet or spreadsheet and track the major forex pairs. If you can at least eliminate the weakest trend patterns, you will be a step ahead.
Same principles for every strategy. You must know what the ideal looks like so that you can identify how far off ideal is your actual situation.