That's a good point. Since it's looking at offenders and not offenses, I don't think they can diverge too much (but perhaps I'm not thinking of the math correct

That's a good point. Since it's looking at offenders and not offenses, I don't think they can diverge too much (but perhaps I'm not thinking of the math correctly). I did find a study looking at just first-time offenders, which found a similar level after five years: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3095888/figure/F1/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3095888/figure/....