On Friday, May 24, 2019 at 8:58:07 PM UTC-7, sms wrote:
On 5/24/2019 3:36 PM, Duane wrote:

snip

Jay is correct. Risk is different for different riders and even for the
same rider in different circumstances. All cycling is not the same.
Sometime I’m trudging through the traffic alone commuting to work.
Sometimes I’m out in the country in a group pushing it. Both have
different parameters regarding risk etc. Some cyclists do only one or the
other. Some do both. Hard to group them together statistically. And what
would be the point anyway?

The point is that when the statistics don't support your premise you
have to do something to justify your position. No one would think any
worse of Frank if he used statistics honestly.

Are you capable of using statistics properly? Hardly anyone is. I have had to use these things in engineering settings where PhD physicists with years of math and statistical analysis under their belt couldn't properly identify problems hidden in the statistics. Being able to see that and correct for it is how I remained at so many important positions.

Even a slight problem - Tesla called to interview me for a job. I told them that I didn't think that they should call their feature an "autopilot". The man hung up on me in a huff. ANYONE should have been able to see the statistical relevance - any accident in which that "autopilot" was engaged automatically became the fault of Tesla. Therefore the insurance companies would tell their customers to always have it engaged. Tesla now call it a "navigation feature" and you are required to have both hands on the wheel when it is engaged. That sure was a difficult one to see coming.