genec

John, it has nothing to do with determining "the motorist who will hit you from behind from the motorist who will steer past," and everything to do with finding a gap in fast moving traffic.

I find a mirror quite handy when I am merging with 45MPH traffic, and need to find a gap in that traffic while I am also keeping an eye on the road ahead. When I spot a suitable gap, I can then turn my head to confirm the gap... and offer a subtle signal that I am moving to the left. If I can, I will then also offer a proper hand signal. Then I will make my move. In dense, fast traffic, the mirror makes all the difference in the world.

Watching for the motorist "that may hit you" is impossible.

Helmet Head

First, we agree that the probability that any one given cyclist will ever be hit from behind is certainly very small. Compared to the probabilities of harm coming from ahead, it's fine to ignore it. Never-the-less, the probability is not zero, and if a practice can reduce that non-zero probability to something significantly closer to zero while helping with improving one's safety with regard to hazards up ahead too, why not adopt it?

I agree that if you wait to try to determine whether a given driver is going to steer into you or miss you until the point where he is just a second or two back, then the probability is extremely small that one will be able to distinguish the motorist who will hit you from behind from the motorist who will steer past. But that's not at all what I'm suggesting.

It's all about "reading" traffic behind you long before they've reached you, and, in particular, distinguishing the potentially distracted ones from the ones who have noticed you.

Of course, the whole thing depends on the assumption that the guy that you have confirmed has noticed you, will not steer into you, but I think that's a reasonable assumption.

Helmet Head

That's the convenience factor that he referred to is, and he's fine with it.

Certainly if you're thinking of a line of traffic passing you by, and the possibility that anyone in that line might suddenly drift into you out of the line, yes, that's impossible to detect in time.

Of course, that's not at all what I'm talking about.

I'm talking about the situation where there is no other same direction traffic at the moment, and someone is coming. I think you can do a lot to determine if that particular driver has noticed you. Further, I think a driver in that situation is about the only kind of driver who will drift and hit a cyclist, so that's where the focus needs to be, if that's what you're looking for.

genec

I still don't buy your latter use, honestly. I toured for years and did not have a mirror most of the time then... and even now when I am out on quiet roads where those long gaps do exist... I can't determine enough about an approaching motorist from any distance with a mirror to make any difference. Usually I hear them long before I might "in a glance" notice them... and I certainly cannot determine anything such as "did they see me" for any perceivable distance... especially when they are moving over 50MPH.

John Forester

You agree with my point. I have never opposed using a mirror for convenience, and, as I wrote, I now use one because my neck is a bit painful.

Helmet Head

I certainly agree that if I'm riding in the margins, I can't tell.

Try riding in the center of the lane and see if you can tell whether they do anything that tells you if they noticed you or not.

For example, picture the climb out of La Jolla on Torrey Pines Road. After the fire station, there is a straight part, with onstreet parking, then a curve to the left (at Pottery Canyon), then another long straight stretch. On that second long straight stretch I am usually out of the bike lane (even the nice new one) and out in the middle of the lane. I could be nearing the top of that straight stretch (but before it starts to curve right around the big sweeping turn to the right through the eucalyptus grove) and see motorists still a long way back, perhaps 10 or 15 seconds (at 50+ mph). From that far back, if I'm in the middle of the right lane, they will often change lanes. Now, if there is a dense pack coming, that doesn't matter much, because the next guy is still coming. But still, I maintain course and wait until he's about 5 seconds back, and noticably slowing a bit, before I signal right and merge right into the bike lane.

Using this technique, I find I can almost always tell if they noticed me or not before it's time to move aside.

noisebeam

I can absolutely tell using glances in mirror is fsdt has noticed me.
Maybe its more important in faster denser multilane traffic vs. rural where one can hear better.
Al