Where does your paper mention a 2 degree downward angle? Your charts indicate a slight upward angle.
In any event it doesn’t matter much. A fall of 5 feet requires .56 seconds so the average speed will be 250/.56=447 fps. And let’s say it leaves the asphalt at 600 fps and when it strikes the manhole it is down to around 300 fps. At 600 fps leaving the asphalt it has (600/2000)^2=0.09=9% of the energy it had before hitting the asphalt. So it lost 91% of its 1860 J=1693J. in striking the asphalt. That still results in a lot of energy transferred to the asphalt.
Where does your paper mention a 2 degree downward angle? Your charts indicate a slight upward angle. The below image is from the report. I posted it here in my first reply to your response. I pointed out the -2-degrees angle in my reply. I drew a red arrow on the graph to make it easy to find.

These graphs are designed to aid in sighting in guns and their sights, and for calculating hold-overs. They are not really made for ricochets. This is an improvised attempt at using one of the graphs for the purpose of demonstrating the idea of the other single bullet theory. I agree that if you are only looking at the graph itself it appears level, but that is not the angle that was input. This is apparently a limitation of the graph. Also, I may need to adjust for only a 5-foot drop, the jury is still out on that. I appreciate your input. I am still experimenting with actual results of actual ricochets.
You are still ignoring the fact that the vertical component of the direction (-2-degrees) gives the ricocheted fragment a “running start” compared to dropping a bullet from the same height.