That's the trouble, you are worried about precision, not accuracy...
Well, if you wish to get technical (I am a scientist, after all), then sure...
The 3 and 5 look about the same in terms of oof-ness. However, it's subjective due to two things: The paper bleed and the CA... Take your pick as to which is going to screw up your judgement more.
Also, I'm on the fence about using something so high-contrast, as this. It's going to introduce CA in just about most lenses that are not of L quality. I'm leaning toward the battery test, in this instance.
Either spend the $2 for a pre-printed chart or print one out on a laser writer.
Not sure how precision comes into this. It's one shot. Precision has to do with repeatability. Humans are bad, at that.
Edit: However, without precision the test is worthless. If the shot is going to vary, from shot to shot... That's why it is important for the subject to be precise. If the CA is different from one shot, to the next, your subjective measurement of it will also likely be imprecise. Your measurement is only as good as the tools used... And, the chart is a tool.