It depends on what you'd like to learn. Particle physicists learn a lot from 'minbias' data, data without selective triggers. If you have a specific question y

It depends on what you'd like to learn. Particle physicists learn a lot from 'minbias' data, data without selective triggers. If you have a specific question you'd like to ask, then selective cuts on the data are useful. If you'd like to characterize an entire system, like Twitter, it's good to randomly sample the entire thing. Without a random sample, you've added your own preconceptions about what's to be learned.