On Sunday, September 9, 2018 at 7:31:12 AM UTC+1, John B. Slocomb wrote:
On Sun, 09 Sep 2018 00:02:17 -0500, Tim McNamara
wrote:

On Thu, 6 Sep 2018 16:24:45 -0700 (PDT),
wrote:

I take that to indicate that a large number of bicyclists deaths are
the homeless or nearly homeless, drunk or nearly so, and riding
probably at night with no lights.

Or the driver of the motor vehicle was drunk or nearly so.

As for homeless/nearly homeless and/or riding at night without lights,
you are reading into the small amount of information Jute provided.
Perhaps there is more detail in the NHTSA link he provided, which I have
not yet looked at.

A number of surveys of bicycle accidents indicate that as many as
half, or more, involve the cyclist disobeying one traffic regulation,
or another and reports from autopsies of cyclists killed in accidents
showed that (in New York) as many as 21% had been drinking (6% of auto
drivers who were involved in a auto - bicycle accident had been
drinking).

http://www.sandiegouniontribune.com/...v22-story.html
https://cityroom.blogs.nytimes.com/2...d-bike-deaths/

It might be noted that these reports are not of the "Well, it is
estimated..." or "It seems as though...", but are statistics, for
example:

"in 2011, officers determined fault in 701 crashes between a bicyclist
and a motorist in which a cyclist was hurt or killed, according to the
reports, submitted to California's Statewide Integrated Traffic
Records System. Cyclists were found to be the party most at fault in
390 of those crashes, or 56 percent of the time, the records show."

It seems odd, to say the least, that these facts, and facts they are,
are never mentioned by bicycle safety advocates.

As Pogo said, "we have met the enemy... and he is us."
(at least 56% of the time)

Exactly. I would regard these "The dead cyclist was drunk" reports with some suspicion when they come from a motorist's mouth, even when I accept other data from the same report. The cyclist, being dead, is by definition not there to deny the accusation, and the motorist may use it as exoneration. On the other hand, a pathologist's report that the cyclist showed on analysis a particular blood alcohol dose falls clearly in a more reliable class of data.

Andre Jute
Statistics sooner or later becomes bias confirmation, but only after people changed their minds on hand of statistics. -- Andre Jute, Chairman's valedictory to MASA