To answer your question, I don't gauge my speed because speed is not what matters. You'll note that when talking about the ghosting method, at no point do I ever mention "go fast" or "go slow". I talk about confidence and hesitation. This can occur at any speed, though generally students will draw more confidently when they draw faster, and will hesitate more when they draw slower. But at the same time, drawing faster tends to reduce control.

The key is to find what speed you can currently ghost and execute at (these are going to be the same speed) without hesitating, and without trying to steer the line with your eyes. With more and more practice, that speed decreases, as we get more, for lack of a better word, confident at drawing confidently.

Anyway, I feel like all this is moot - because looking at your linework throughout your challenge here, you are drawing your lines confidently. So perhaps you're just overthinking things.

As a whole, I am quite pleased with your work throughout this second attempt at the challenge, and it clearly shows just how much you're capable of, and how little room you've given yourself in the past to really demonstrate it to yourself.

Starting with the cylinders around arbitrary minor axes, you've done a great job of aligning your ellipses, have clearly played with rates of foreshortening that were more dramatic (not exactly that much variety still, so in the future mix it up with both shallow and dramatic, both having vanishing points but at different distances from the cylinder), and have done an excellent job of keeping the shifts of degree and scale from end to end consistent with one another.

Moving onto your cylinders in boxes, I did notice that you didn't seem to actually draw the side edges of your cylinders, and merely drew ellipses in opposite planes of your boxes. That is... technically not a huge deal, but is a strange thing to leave out all things considered.

Another thing I did notice was that in quite a few boxes, like 79, 82, 83, 84... a lot, in fact, you have a tendency to draw the far end at an entirely different angle (converging with different vanishing points) from the closer end. This in turn results in ellipses that are narrower on the far end, which as we know about cylinders, is incorrect. Just something to keep in mind.

Anyway, other than that you're moving in the right direction, and I'm still very pleased with your work, especially compared to what you submitted last time. As such, I'll go ahead and mark this challenge as complete. Just be sure to continue practicing your freely rotated boxes, and watch the tendency to set the far planes at an angle.