The short answer is: sometimes yes, sometimes no. I realize that's not too helpful, so let me expand that a bit
First, the basic rule is to diagram anything only when you feel it will be helpful to you. I describe that thinking here: LSAT Conditional Reasoning: When To Diagram
. That goes for all diagramming, not just CPs, and what it means is that what you diagram may change over time. LSAT beginners tend to diagram quite a lot, whereas more experienced LSAT takers diagram a bit less. Each person finds the level where the actions they take help them out.
With contrapositives, I tend to diagram them only when the original rule is not a simple one. So, an original statement such as A
B is clear enough to me that taking a CP and writing it down doesn't help; I already see the meaning of the CP inherently when I glance at the A
B representation. On the other hand, a rule such as "If any side has exactly one of its three lights on, then that light is its center light" (which is a rule from a real LSAT game) would have an interesting initial diagram and I would for sure stop and consider the CP, and diagram it. In other words, if I think there's going to be value from looking more deeply at it, then I will stop and consider it and probably draw it out. that not surprisingly happens more often with complicated rules than simple ones (but not always), so I don't take a hard rule on this, and let context help guide me.
Please let me know if that helps. Thanks!