If the question is whether or not there is a high chance of someone committing a violent crime caused by playing video games, then you are correct. If the question is whether there is any causation link between violent video games and committing real life violence, then you need more data than what you are using.

We have roughly 380 violent crimes per 100k people each year in the US. Just to keep it simple, let's assume they are committed by unique people. Half of adults play video games and about half of those play violent video games. (Very rough numbers but the right ballpark.) So in 100k people, you have 25,000 who play violent games of some sort. Of those violent video games, if only a certain type can create violence, you may be looking at only 10,000 people or fewer (per 100k people) who are playing the games that could cause violence. It also seems likely that the ability to cause violence would be tied to genetic and environmental issues in the person playing the video games. On top of that, for the violence to manifest itself, the person would need to find themselves in a circumstance where it could manifest. (Being bullied at school, etc.)

It seems very likely that by the time everything that might have an impact on violence is considered, you might be only dealing with 1,000 people out of every 100k who would be in the at risk group. If the effect for the at risk group is very high, say 1 out of 1,000, you are talking about one more violent crime per 100,000 people.

It would seem that an increase like this could easily be hidden due to other programs to reduce crime like cleaning up lead pollution, or better training teachers to to identify at risk students.

(And please understand, I'm not saying that there is a link. Just pointing out how it would be likely to show up if there was.)

> Half of adults play video games and about half of those play violent video games

The outrage was mostly about children and permanent damage to their brains caused by games. Almost all children play games, and certainly more than half of all games (weighted by popularity) are violent.

> very high, say 1 out of 1,000

That seems the opposite of high. High would be 1 in 2 or 1 in 5.

If the effect of everyone playing video games is 1 additional crime per 100 000 people I don't see the problem TBH.

I get what you are saying. Keep in mind I'm just saying this would be how we look for causation, not whether it makes sense to actually do something about it.

> That seems the opposite of high. High would be 1 in 2 or 1 in 5.

I guess I'm thinking of how we'd view a side effect in medicine. If 1 out of 1,000 people who had the problem the medicine was supposed to cure were likely to die or cause a death when taking the medicine, it would seem like a high number of deaths. You'll routinely see side effects of things that happened 1 in 50,000 cases being listed as possible side effects of medicine.

> If the effect of everyone playing video games is 1 additional crime per 100 000 people I don't see the problem

Whether this matters probably depends on what violent thing they do. If it is an additional Columbine type event for every 100k people, that would be much less acceptable than one additional instance of verbal abuse by someone in a disagreement at a bar.

In medicine they give the probability out of all people using the drug, you gave the conditional probability with many assumptions.

This would be like saying "this drug might kill 1 out of 1000 people who also have this rare gene which only 1% of population has".