I think there's a big difference between "everyone else is awful", "there's statistically enough awful individuals" and "every driver is trying to kill me". Only the last one of those is going to lead me to incorrectly blame a whole group.
I agree that a statistically significant subset of drivers cause problems (and a statistically significant subset of cyclists, too), but the vast majority of individual drivers (and cyclists) are safe and do not deserve to be grouped in with the the problem-causers.
They have to be grouped in, as there's no way to differentiate them.
Also, you're proposing a false equivalence. When drivers cause problems, other people (pedestrians, cyclists, other drivers) die. When cyclists cause problems, they're usually the only ones that get hurt. The urgency of the two problems is dramatically different.
I agree that a statistically significant subset of drivers cause problems (and a statistically significant subset of cyclists, too), but the vast majority of individual drivers (and cyclists) are safe and do not deserve to be grouped in with the the problem-causers.