A pose "Reads" correctly only if the audience, looking at it, can tell at a glance what the entire physical position of the character is. Even better, of course, is if the pose puts across a strong sense of motion or anticipation, so that the audience is cued to understand the next frame of the animation even before they see it.
How well a pose reads is often measured by how well its silhouette reads. In general, the only thing that the audience is guaranteed to perceive is the silhouette of the character. You can hope that they will pick up on more subtle visual cues, but you can't count on it. Honestly, even if they do pick up on subtle visual cues, these will be reinforced by a good silhouette on your image.
A
strong silhouette will make everything important about the action clear.
It will show the positions of all four limbs, the way that the character
is bearing their weight, and the general focus of the character's attention.
A
bad silhouette will obscure as much as possible. It will cover the
positions of the limbs, or make it impossible to tell (by silhouette) which
limb is which. It will hide the way the character is bearing their
weight, and make the focus of attention a mystery. A really truly
bad silhouette can reduce a well-modelled character into a featureless
blob of color and lines.
How well a pose "reads" depends not only on the pose, but on the angle
of the camera, other characters, the color of the background, the nature
of the lighting, and probably other factors that I haven't even thought
of. You should try to keep all of these things in mind... of course,
you probably won't. Certainly I never do. But the more you can be
concerned with giving your character every chance to shine, the better
your animations will be.