Some information you get from an experiment is not 100% accurate (for obvious reasons) but theoretically should produce consistent results. Because data can vary from what it's 'supposed' to be, we draw a line of best fit to essentially 'summarise' our data, and produce patterns that are more...