To determine which graph best models the data without any calculation you have to look at the relationship between the dots and the line to see which one fits better.
The dots represent the observed data, the values that apear in the table.
The line that best fits the observed data will be that one that is closes to the dots. This means that the best of fit line will be the one for which that passes through the middle of all points, and pass through most of them.
The line in graph A passes through the miffle of all points and has two of them over it, the distance between the dots and the lines seem to be the shortest in this graph, so this one is the one that models the data better.