In order to stop engaging in discussions about "using DLLP to avoid inadvertent drift collisions" in every thread in which it seems relevant, I would like to devote this one thread to that topic, and stop engaging in this debate in other threads.
I will start by presenting the theory (a "hypothesis" in scientific terms). This includes some definitions of terms that I use, the premises the theory is based on, the reason, conclusions, and some FAQs. I hope this helps.
- ruralish - any non-urban road with long stretches of no intersections.
- DLLP - Dynamic Lateral Lane Positioning - a cycling methodology that emphasizes use of lateral lane positioning for improving the cyclist's sightlines to potential hazards, conspicuity and predictability.
- look back - A clear and obvious look back over your shoulder for longer than "just a glance".
- merge - moving laterally in a safe and predictable fashion... for a bicyclist that means only after looking over the shoulder on the side on which he plans to move in order to verify it is clear and safe to make the move.
- weaving - to move back and forth between two lateral positions with a period of less than 10 seconds.
- fog line - a stripe at the edge of the road that vehicle drivers are not supposed to cross.
- shoulder - any paved portion of the road to the right of the fog line.
- bike lane - a shoulder that is designated by signs and/or markings to be a "bike lane".
Premises (to determine the validity of the argument, these are assumed to be true; to determine the soundness of the argument, it must be valid and these premises must be accepted to be true):
- In a "ruralish" (non urban) environment on a quiet low traffic open highway with good sight lines, a potential obstacle up ahead in a motorist's intended path in his lane is much more likely to be relevant to that motorist than is the same object up ahead in the shoulder or bike lane.
- We are much more likely to notice things which are relevant to us in some way.
- If you're more likely to be noticed, you're less likely to be overlooked.
- A contributing factor in the type of inadvertent drift collisions cited in the bike lane deaths thread is the motorist overlooking the presence of the cyclist
- Another factor in those inadvertent collisions is the motorist choosing to attend to a distraction that he probably would not choose to attend to, or would not choose to attend to as long, if he had noticed and was aware of the cyclist up ahead.
- The inadvertent drift collisions occur when the distracted motorist drifts into the shoulder or bike lane while attending to a distraction.
- On long straight stretches of road, motorists who are following someone else relatively closely (i.e., gap to vehicle in front of them is less than 5 seconds) are much less likely to choose to take their eyes off the road for a long enough time to drift significantly out of their path than are motorists who have an relatively empty road (intended path is clear from vehicle to point that they won't reach for at least 5 seconds) in front of them; the longer the gap, the more likely they are to take their eyes off the road to attend to a distraction for a long enough time to significantly drift out of their path.
- There is no legal requirement for cyclists to "keep to the right" when same direction traffic is not present.
- On most ruralish narrow roads with no shoulders, it is safe to cycle there even though doing so requires every approaching motorist to notice the cyclist in order to not hit him.
- Per A, a cyclist up ahead in a motorist's intended path in his lane is much more likely to be relevant to that motorist than is the same cyclist riding up ahead in the shoulder or bike lane.
- Per B, the cyclist that is more relevant is more likely to be noticed by the approaching motorist.
- Per C, the cyclist that is more likely to be noticed is less likely to be overlooked.
- Per D, E and F, the cyclist who is less likely to be overlooked is less likely going to be inadvertently drifted into.
Per the above reasoning, it's safer to be up ahead in the motorist's lane than up ahead in the shoulder or bike lane: you're more likely to be noticed, and therefore less likely to be inadvertently drifted into.
Therefore, what I often do when there is no same-direction traffic is merge left into the main traffic lane, and stay there until I notice with my mirror faster traffic approaching from behind. When they are about 10 seconds back, I do a right shoulder check and move back into the shoulder or bike lane.
- Why would you want to weave back and forth in and out of the lane like that? It's not weaving. The technique normally involves maintaining a given lateral position for at least 10 seconds, almost always much longer, before moving back to the other position.
- If you're in the shoulder or bike lane being passed by cars, and then there's a gap, what is the minimum that gap has to be in order to move back into the traffic lane? I will rarely move back into the traffic lane unless I can see that the gap to the next vehicle approaching from behind is at least 20 seconds. If it's less than that, I might move out to the right tire track for a few seconds, then move back, just to increase the chances of them seeing me, demonstrating look backs and clear arm signalling for each merge.
- Would you do this without a mirror? Only on quiet slow roads. On roads with fast traffic (35+ mph speed limits), I would not do it without a mirror.
- How does a shoulder that is safe for cycling while being passed by motorists become less safe when same-direction faster traffic is not present?. During the gaps in same-direction traffic, the next motorist to be approaching is more apt to paying less attention to the road (see Premise G above), and, therefore, more likely to take his eyes off the road long enough to drift into the shoulder or bike lane. That's why it's less safe to be in the shoulder during gaps in same-direction traffic.
- It seems like a lot of work. Is it? You say "work", I say "paying attention", and that's a good thing. In a shoulder or bike lane it's all too easy to lapse into a mindless wandering of thoughts where we can become as oblivious to traffic as some of the overtaking motorists are to our presence. That's not good. Staying engaged in DLLP helps keep the cyclist alert and paying attention to what is going on around him. That's good. It's good to notice in your mirror that there is no same-direction traffic and to move out into the main traffic lane, and to pay attention with your mirror (a microsecond glance every 3-5 seconds all it normally takes) to note when traffic is approaching from behind, and to move out of the way when you've been out there long enough in their path for them to have been very unlikely to still not have noticed you.
- If a driver is not paying attention, how is he going to notice you regardless of where you are riding? Drivers periodically and frequently, if not constantly, pay attention to their intended path, or they would be driving off the road all the time. The point of this technique is to do be out their in their intended path long enough so that it would be virtually impossible for them to notice you, because that would mean they were not paying attention to their intended path for so long, there is no way they could stay on the road.
- But this technique is supposedly about avoiding "inadvertent drift", which is exactly that - drivers not paying attention to the road so long that they can't stay on the road. It's a timing thing. What you want to do is try to get their attention when they're still so far back (more than 20 seconds), that if they're already not paying attention at that point, they'll drift off the road and crash long before they reach you. Odds are, that that's not going to happen, and one of their periodic/frequent checks on the road ahead is when we're counting on being noticed. If, instead, during one of those checks, we're in the shoulder, then we're much less likely to be noticed. That's the point of this technique: to be in the space where the motorist is paying attention when he's paying attention..
- But since some motorists are not expecting to see a cyclist in the road up ahead, aren't we prone to inattentional blindness when we're in the road? First, the expectations for cyclists in the road is probably not any signficantly different on the narrow roads, yet they don't seem to hit cyclists on those roads because of inattentional blindness. Second, the technique involves moving aside into the shoulder to be considerate and to accomodate their easier passing of us. In the highly unlikely event that they haven't noticed us despite us being in their path, it doesn't matter, because we're moving out of their path anyway (as opposed to what we do on narrow roads - stay in their path). Finally, the likelihood of the motorist not seeing the cyclist in his path, combined with the low likelihood of the DLLP cyclist not noticing the approaching motorist in his mirror, and so staying in his path and not moving aside, becomes an extremely unlikely event. See the bike lane deaths thread for examples of much more common (and therefore more likely) events.
- The difference in angle of vision between a cyclist up ahead in the shoulder and a cyclist up ahead in the traffic lane, when the motorist is 20 seconds back, is so light, how can one cyclist be more noticable than the other? The argument has nothing to do with angles (which has to do with sensory conspicuity) and everything to do with relevance, a key aspect of cognitive conspicuity. That's not to say that relevance is required to be noticed, just that relevance makes something much more likely to be noticed (see Premise B). What makes a cyclist up ahead relevant to motorist is not whether he can see him (because we're assuming he can see the cyclist), but where the cyclist is relative to where the motorist is going (or at least thinks he's going). It is normal and natural for the subconscious mind to assume that a cyclist in the bike lane or shoulder is going to continue riding in the shoulder, and that the motorist will not enter the shoulder, and, therefore, the cyclist is not relevant to the motorist.
- Are you suggesting the primary reason one should move left is to improve conspicuity? - Robert Hurst No. See explanation in Post #66.
I reserve the right to update the presentation of this theory in this OP, but will always note edits here in this section, and, when signficant and appropriate, will make a post about the change.
- Added FAQ #5, #6, #7, #8, #9
- 10/1/06: FAQ #10; revised description of what "Premises" are from must be taken on faith to must be assumed to be true.
- 10/2/06: Updated premises description again.
- 10/4/06: Added sightlines to DLLP definition.