Abortion? Not very convincing. In fact, when other researchers took a stab at that hypothesis, and directly compared the crime rate against age cohorts, they found a statistically significant increase in murder rates. The timing showed the highest increases amongst young males after the legalization in 1973 at the time when the crime rate should have been dropping. Considering the age of young males is central to their hypothesis predicting the timing of the drop in crime rate, that doesn't lead to very strong evidence in favour of that hypothesis.
The correlations don't hold when age is a predictor.