3 ms·
Did you see the light turn yellow? How long does this light stay yellow? Is this the first of a sequence of timed lights? Is the intersection photo-enforc
by dmckeon 14y ago
Did you see the light turn yellow?
How long does this light stay yellow?
Is this the first of a sequence of timed lights?
Is the intersection photo-enforced?
Can you enter the intersection before the light turns red?
Will speeding up get you into the intersection
before the light turns red, or after?
Do you carry comprehensive insurance?
If a collision ensues, will you be at fault?
I doubt you will find a mathematical proof that is usable in actual
traffic, but a statistical approach might be usefully informative.
Have you measured the total possible minimum and maximum wait times
for all light combinations over some of your usual trips?
(Hint: get your friends to do the field work.) How do the "all red"
and "all green" times compare proportional to the total trip time?
What are the chances that waiting for one yellow light will increase
the typical total trip time?
If none of that convinces you to wait for traffic lights, then take
the MSF (Motorcycle Safety Foundation) course, buy a small
motorcycle, and report back after a year of riding in traffic.
The experience may give you a different perspective.
- 31reasons 14y agoYou probably didn't read my whole post. I don't speed up at yellow light like most people do and I don't ever intend to. What I am interested in is a theoretical insight into proving using graphs/probability theories with most simplified assumptions. I don't intend to do the field work, it could be better done with computer simulation but I am not sure if thats the better way to approach the problem. This is a theoretical question only, no MSF involvement necessary.