scott preece on 55mph
Steven M. Haflich
"the statistics I have ... "
What statistics? Where from? It wasn't even clear from your posting
which conclusion you were drawing:
1) The same number of people will be killed if more or less
everyone drives 75 as 55, or
2) Chances of being killed in a crash are about the same if
one crashes from 75 as from 55.
The first "conclusion" might stand somehow -- arguments about number
of driving hours increasing have already appeared on the net -- although
I doubt it. The second "conclusion" is very strange. You doubt that
the curve relating liklihood of injury/death per accident and speed
is smooth? Whatever the shape of the ciurve, it is unlikely to be
particularly discontinuous! The question is whether there is a
significant difference between 55 and 70 or 75. Again, the old
engineering terminology -- "finding the knee in the curve" --
comes to mind.
You might at least make clear which of the two assumptions you
are flaming^H^H^H^H^H^H^Hsupporting (sic). But remember, the
distance you travel after deciding to panic stop (or whatever)
goes up (in first order approximation) with the SQUARE of speed
(or rate of velocity, or speed of rate of integral d^2x/dt^2, or...).
Try typing "scale=4 \n (70/55)^2" at bc some time.
Steve Haflich
genrad!mit-eddie!smh
la...@utcsstat.uucp
an estimate), and speed limit, and number of deaths for the Provinces of
British Columbia and Ontario and the States of Ohio, North Carolina and
California from the years 1960-1981.
They show that the number of deaths is very much proportional to the
number of cars.
They show that there was no significant lessening of deaths due to
the 55mph speed limit change -- all the reduced death figures can
be accounted for as proportional to the reduced number of cars on the road
the year that 55mph speed limit went into effect.
The cheaper gas is the more people drive the more traffic congestion
there is the more people get killed at any speed over 40 mph.
I posted all this weeks ago. I have been arguing ever since. I figured
everybody had heard me say this before and wasnt interested in hearing the
*whole* argument again. The first time it took me 200 lines.
I guess I was wrong.
If you plot speed to deaths for the above mentioned places for any year
you get a graph WITH DISCONTINUITIES IN IT. 30mph is right smack in
the middle of the most noticable discontinuity, but there are others.
Almost nobody gets killed if the speed limit is 20mph or less, except
that there is a sharp peak at 15mph. Lots of people get killed at 15mph
but nobody seems to at 20.Why? I have a theory. there are a lot of
"dangerous curves" in Ontario which are marked 15mph (except that we do
it in kilometers now). People miss the turns and go over escarpments.
There is no way to know exactly what speed they were going, and they
fall into the statistical group for the 15mph even though they may
have been going 30mph. Since some of the most dangerous spots are
marked at 15mph it does not make no sense to see that more people get
killed there than on safe city streets at 30mph.
Barring the 15mph peak, you see that very few people get killed
until you are travelling in a 20mph zone, then there is a huge
jump, and people start getting killed at a much higher rate at
somewhere above 40mph. Adjusting for number of cars on the road
you get a nearly flat line from 40mph up until you reach 85mph, which
was the highest (legal) speed in any of the areas I mentioned.
Always adjusting for the number of people on the road, you get a
flat line until 20mph, with a big spike at 15mph, a sprinkling of
figures, and then another parellel flat line starting at 50mph.
In between you cannot draw a curve. You dont nearly have enough
figures, since no government is going to have a 47 mph speed limit
so I can collect more figures. But it is interesting to note that
from year to year the death-per-100-drivers-adjusted-for-volume-of-traffic
for speeds between 20mph and 55mph varied enormously. One year, 25 mph
was the big killer in BC while other years showed 40mph to be the
dangerous section.
conclusion? there is no one-to-one correlation between speed and
driving deaths, beyond and understanding that at a certain speed or
higher you are going fast enough to get killed and at another certain
speed or lower, in general, you arent.
There are other interesting things you can discover. When BC made more
paved roads out of old gravel ones less people died. When Ontario did
the same thing, more people were killed on the highways than had been
killed when the roads were all gravel. The reason? There was some hazard due
to having gravel roads (where you can skid). In Ontario, the increased use
of the paved roads outweighed the reduction in hazard due to gravel. In
BC it didnt.
Laura Creighton
utzoo!utcsstat!laura
ps Now folks, what do I do? Faithful readers of net.flame and net.auto
have heard all this before. I said it already. I will get hate mail about
duplicate articles and hate news. But if I dont go over and over the same
things again and again other folk dont understand the ongoing focus of my
argument abnd say that I havent expressed myself clearly. Who do I burn,
folks? The people that are sick of reading me restate myself, or the people
that have forgotten (or never read) what I have already said?