I would guess that at least one out of four drivers is exceeding the speed limit at all times, so the fact that one out of four fatal crashes involves speeding

I would guess that at least one out of four drivers is exceeding the speed limit at all times, so the fact that one out of four fatal crashes involves speeding doesn't say much. it's odd to start off the article with such a weak statistic and say nothing about how tenuous the causal link is, but the the rest of the article does a pretty good job exploring the absurdity of US speed limits.