I do understand that; buzzman was just complaining about all the "asterisks" I had included in my risk compensation argument, so I'm not sure how you could see that as taking it as "holy writ". Some are arguing we should throw it out altogether because it's effect, as it applies to cycling, is not well studied. I'm arguing that it's a consideration to me made in the assessment; not that it's a trump card.

*sigh* I'm not ignoring data. Yes, there is contrary information, just as there is for pros/cons of helmet efficiency in the first place. Even if you don't like the ABS example, there are others out there. It simply cannot be denied that when the risk/reward assessment we make changes, it affects decisions we make based on that. I like to use the teen example in this case because it was fairly well studied in the mainstream, not by outliers in a small field like bicycle helmet research. The concept is the same. Again, what extent this applies to cycling, specifically helmets, can't be answered thoroughly because it hasn't been well researched; but we know it must be a factor due to our more extensive research on the risk/reward assessment in general. Perception of safety and risk matters a lot.

Imagine if all cars were made of Nerf and accidents were harmless; do you really think people would drive as "carefully"? Of course not, because perceived risk has dropped. This is an extreme example, but it demonstrates the point. The same thing happens with safety gear. It's more subtle, and varies based on how people perceive it affects their level of risk, but it's the exact same assessment process we all make, both consciously and otherwise.