> I'd suggest that unless you have very atypical habits that put you in conflict with bicycles very often and cars very rarely, you are worrying about the wrong

> I'd suggest that unless you have very atypical habits that put you in conflict with bicycles very often and cars very rarely, you are worrying about the wrong thing. While you would be correct if only considering risk of death, this doesn't tell us anything about the incidence of non-fatal pedestrian-cyclist interactions and what the outcomes are. It would be perfectly rational to care more about cyclists if it was 100x more likely than a car crash and still carried a signifigant risk of injury, even if the risk of death is extremely low. To be clear I actually agree with you that cars are still likely the higher risk, just pointing out a limitation in your data.