This is a little bit of a frame challenge, but I don't think that it matters all that much. What is the graph of a line trying to convey? What data are really necessary to get right? In my opinion,
- an approximation of the slope (very negative, slightly negative, flat, slightly positive, very positive),
- a general indication of the location of the intercepts, and
- there is no (3).
My general advice to students when graphing these kinds of things is to either (a) sketch two easy-to-find points and connect-the-dots, or (b) sketch one easy-to-find point and use the slope to finish the graph. "Easy-to-find" is a bit of a dodge—it doesn't really explain what is going on—which is why I like to give a few random examples of different kinds of lines, generally by using a set of 20-sided dice to determine coefficients (plus a coin to determine the signs of coefficients). What is "easy-to-find" is going to be down to taste and experience.
In the example given, I think that the intercepts are relatively easy-to-find, so I would expect an answer from a student which looks something like the following:
My students will often work very hard to draw everything to scale, but you'll notice that I haven't bothered. The important information is there: the exact location of the intercepts (more importantly, the $x$-intercept is on the positive $x$-axis, the $y$-intercept is on the positive $y$-axis, and the $y$-intercept is closer to the origin than the $x$-intercept), and an indication that the slope is slightly negative (i.e. negative, but not super-steep, i.e. somewhere between $0$ and $-1/2$, perhaps)
I think that the philosophy that students should be taught early is that a graph should show key information, and that everything else can and should be left a little imprecise. This is true for lines, for parabolas, and for all the curves graphed in calculus and beyond. If one needs a really pretty picture, computers are very good at making super accurate graphs—the point of doing things by hand is to identify important features, not to get a publication-ready image.