If 95% of cyclists wear helmets (that would be my guess, at least when cycling on the road) and only 80% of fatalities happen with a helmet then it would seem t

If 95% of cyclists wear helmets (that would be my guess, at least when cycling on the road) and only 80% of fatalities happen with a helmet then it would seem to me that helmets are very effective - which is why you need to qualify the statistic.