You are not wrong generally about potential HN bias, but to be fair, the onus is on the people who believe video games are causative factors of violent behavior to first show correlation. If they can't even do that, then we are under no obligation to throw away the null hypothesis.