What I was saying was that your interpretation of the chart was wrong, not that the chart was wrong. I was trying to illustrate that other factors can explain variations of crime much better than sentencing. What I was saying, for instance, with this chart and the previous chart, was that you cannot explain the variations in crime across Canada by sentencing policy. The fact that violence can be very high in one area and very low in another is not something you can attribute to differences in policing, differences in courts, differences in prisons, or differences in sentencing.
In Canada, we have a single federal Criminal Code. We have very similar policing structures and abilities. If you're going to say, for instance, that in one community, such as Regina, over a ten-year period you saw a 50% increase in violent crime, whereas in another, such as St. John's, you saw a 50% decrease in the same period, clearly that's not attributable to sentencing.
Similarly, variations in important economic conditions, such as rates of unemployment, do produce changes in terms of crime that you cannot identify.... You cannot find similar data anywhere that shows that changes in sentencing policy would do the same.
My point was that if you want to do something about serious violent crime, work on the problems that actually will produce results. Endlessly playing with sentencing will simply not do it. There's just no evidence that it will achieve that goal.