The data set which is best fit for the regression line is Data set C.
Hence, the correct option is Option C.
It is option C because, from the very definition of what residuals are - the the difference between the actual values and the predicted values - we know that the closer the residuals are to zero (indicating a smaller difference between tha actual and predicted value) the more accurately the regression line fits the data. The graph in option C shows residuals which are all very close to zero.