I was confused by this as well. You're comparing the wrong statistic, though: 40k is for all auto deaths; pedestrian deaths due to cars is around 7500. But st

I was confused by this as well. You're comparing the wrong statistic, though: 40k is for all auto deaths; pedestrian deaths due to cars is around 7500. But still, 67 vs. 7500 is still a large gulf. If they're only expecting these new measures to reduce pedestrian deaths by about 1%, the changes don't go far enough at all.