Yes. This kind of micro-A/B testing ("red or green buttons?") feels analogous to premature optimization when coding. Don't worry about the tiny 0.0001% improvements you get from using a for-loop over a while-loop; improve the algorithm itself for order-of-magnitude changes. Focus on the big picture.