You need to show some work here. Look at the extreme points, which look
to be (0, 50) and (10, 950), when rounded to the nearest tens.
Work out the slope of a hypothetical line between these points.
This hypothetical line divides the points, where more of them are below the
line than are above it. None of the answer choices include 50 as a vertical
intercept. However, consider if a "better line of best fit" is achieved by
moving down the hypothetical line a smaller increment. Then it can have
about the same number of data points on either side of it. That would
decrease the vertical intercept by a proportional amount.