I'm pretty sure if you look close you'll see the die polishing lines characteristic of 4A, or at least some of them. The 4A plate coin at VAMworld shows the same clashing as yours does. I believe sometimes "double clashing" can result from a "bounce" very close to that of
Machine Doubling, and minute clash doubling like yours sometimes happens in the same die stroke.
Having been created, those polishing lines immediately start to wear during striking. They're not really deep in the first place, and since we don't know how quickly the die cracks appeared or how long the dies survived with them, they could well have struck enough additional coins that the lines mostly wore away. A new end-of-life sub-VAM being discovered for V4, where 4A's polishing lines are worn away and the coin is back to looking like the original VAM-4, isn't out of the question. It's happened in VAMming before.
Meta-note about the research process for this one: Check first to see what VAMworld gives you before attacking the attribution. The 1887-P
VAM page has a link to VAMs broken down by date position, which immediately laser-focused me on this coin's date as I then knew extensive data was available.
That made me discover the coin is a classic Near Date, and VAM-4 was therefore the second coin I looked at in research. Total time for attribution: maybe sixty seconds.