[frostburg]

This is an abuse of statistics, because you're not considering the likelyhood that a random 15-time felon killed someone, but the likelyhood that a specific 15-time felon that was accused of killing a specific person actually did it.

If you were trying to pick a realistic scapegoat he would have been a great candidate.

[ljsprague]

>If you were trying to pick a realistic scapegoat he would have been a great candidate.

It also helps if the scapegoat had the victims property in his car.

[frostburg]

I have not reviewed the details of the case and have no opinion on actual guilt, I was commenting on a specific line of reasoning.