It's also not helped by the graph for the USA the article uses. Sure, you've got three spikes of race-related events at half century intervals, but all the othe

It's also not helped by the graph for the USA the article uses. Sure, you've got three spikes of race-related events at half century intervals, but all the other time series are uncorrelated (and even for race related incidents there's an obvious structural difference between centuries).