I dont think that data say whatever you want them to say necessarily. If that were true then there would be no reason to collect data ever.
why dont you inspect the data and show us what YOU want it to say. it seems pretty clear that if the vic was white and the def was black as opposed to white, the DA is substantially more likely to go after the DP. That is consistent with the fact the vast majority of DAs are white, dont you think?
That's a big leap. is there no other possible conclusion? could there not have been other factors involved, such as the fact that perhaps those murders committed by the black defendants were particularly heinous and deserving of the DP for reasons having nothing to do with race? i'm not saying they were, i'm just saying that it's possible that there were other things involved.