Reports of the demise of this Web site are greatly exaggerated! We at thank Harris Cyclery for its support over the years. Harris Cyclery has closed, but we keep going. Keep visiting the site for new and updated articles, and news about possible new affilations.

Cycle Sense:
Why Reflectors Don't Work

There are optical reasons for all those crashes and deaths

find us on FB
by John Schubert
HTML formatting by Sheldon Brown
Spoke Divider

Spoke Divider

If you go for a drive tonight, you'll see reflectors shining brightly from mailboxes. You'll see reflectorized stop signs. If bike riders are out, you'll see their pedal reflectors . All these reflectors will appear bright, and very easy to avoid.

So here's the seven-million-dollar question: If all these reflectors are so darn bright and easy to see, how come the bike safety nerds insist you need active lights to be seen at night?

There is a very scientific answer: reflectors work only under very specific conditions. Those conditions happen to prevail in most of the nighttime driving we do, so we get the impression that reflectors work most or all of the time. But reflectors don't work at all if those conditions aren't met, and many well-defined bicycle crash types occur in situations when we can expect reflectors to not work.

Few people understand how easy it is to wander outside the range of conditions in which reflectors will work. But it's astonishingly easy.

Why would a reflector decide to malfunction? And how could it? It doesn't have electrical components to fail, like, say, a British car.

It does, however, have other limitations. Among them:

This list is surely incomplete, but it makes a point: many factors can prevent a reflector from beaming light at the intended observer. This point is not hypothetical-our nightly accident rate shows that. Roughly once per night in the USA, a person is killed on a bicycle after dark. Many more are injured. Very often, I suspect, these accident victims have Federal Consumer Product Safety Commission approved- and required- reflectors on their bikes. So here's my message to those who say, "These reflector requirements are safe and effective." You've lost all credibility.

To understand how reflectors can fail, you need to learn about fifty cents' worth of industrial engineering. The relevant topics are: entrance angle, observation angle, headlight beams, positions on the roadway, the human propensity to make big mistakes, and "other".

Spoke Divider

Spoke Divider

Entrance Angle.

A reflector is brightest if it's perpendicular to the light shining on it. For example, if an overtaking car is directly behind a bicycle, the bicycle's reflector will shine brighter than if the overtaking car is coming around a curve in the road when its headlights train on the reflector.

"Entrance angle" measures how close to perpendicular the reflector is. Reflectors are very bright within about ten degrees of perpendicular. Then, as the angle moves farther away from perpendicular, the reflected brightness drops off in a rather steep bell curve.

In recognition of this steep curve (in the real world, in recognition of the fact that roads have sharp curves, and bicyclists wobble), the Consumer Product Safety Commission (CPSC) required bicycle reflectors to be redesigned and, in their view, upgraded, to meet their 1976 bicycle safety standard-the same standard in force today.


CPSC-spec reflectors have three surfaces at 30-degree angles to one another, so that an approaching headlight will point to a section of reflector with a small (highly reflective) entrance angle through a broad range of car-to-bike angles.

We Fred-types often debate whether the CPSC-mandated design, by carving the reflector's surface area into three small pieces, sacrificed needed rearward brightness to add unneeded brightness from irrelevant rear-quarter angles. I think this discussion is a bit of a sideshow. That's because I believe that a bicyclist needs an active taillight (and is much safer with two taillights, should an overtaking motorist be inattentive or drunk). A bright reflector is nice-but only as icing on the cake.

Another factor to throw into this debate is the latent assumption that more brightness equals more safety. One expert who doesn't think so is Richard Blomberg, who has done numerous studies of these questions under government contract. Blomberg has found, and published, that some dim objects are quite attention-getting. (The old blinking Belt Beacon flashing light scored top marks for detection distance in one of Blomberg's studies.)

Three Freds and their flashlights demonstrate how the entrance angle affects reflector performance. The Fred on the right will see the brightest return from the reflector, since his entrance angle is zero. The Fred in the middle will see a little glimmer of light; his entrance angle is about 30 degrees. The Fred on the left, whose entrance angle is about 60 degrees, probably won't see any light reflected back.

Observation angle.

You ever hold a flashlight up next to your eye and point the flashlight at a reflector? The reflector looks pretty bright. But if you hold the flashlight at arm's length and point it at the reflector, the reflector doesn't look so bright. How close to your eye does the light source have to be to your eye for the reflector to look bright?

This measurement of closeness is called "observation angle," and it's the angle between the beam of light heading to the reflector and the beam bouncing back to your eye.


For a bicycle reflector or typical reflective garment/accessory to appear bright, your eye needs to fall within a very narrow observation angle. And in measuring the performance of reflectors and reflective accessories, the industry has standardized on an observation angle of 0.2 degrees. What's that mean for Joe Driver? I've already done the trigonometry for you; in most traffic situations, Joe's eyes are far outside that 0.2 degree arc.

Reflector performance is are also measured at an observation angle of 1.5 degrees, a figure which is realistic for car drivers. (But even 1.5 degrees isn't realistic for a truck driver, whose head is many feet above his headlights.) But when reflector jocks get together and talk about performance, they talk about the candela they measured at 0.2 degrees. So they're using a measurement which, although true, is irrelevant-and the brightness numbers it produces are deceptively large.

Should the motorist's left headlight be burned out or mis-aimed, then the reflector-equipped bicyclist is now relying on the soberingly large observation angle between the driver's eye and the right headlight.

Here's one reason why a taillight, rather than a reflector, is sometimes essential. A reflector or reflectorized garment may not be visible to an overtaking motorist with a burned-out headlight. Reason: the reflector shines light directly back at the light source. Very little light is reflected beyond an angle of 1.5 degrees (shown here as a 3-degree-wide cone). The motorist, and his eyes, are outside of that cone.

Spoke Divider

Spoke Divider

Headlight beam patterns.

Auto headlights are aimed down and slightly to the right. So bikes on intersecting streets (like the one in Johnson v. Derby), and even oncoming bikes, rely on light scatter - random light that isn't even designed into the auto's beam pattern. Usually, the scatter is bright enough to make reflectors in the oncoming lane shine back at you. But not always. It's a bad idea to trust your life to scatter.

Headlight beam patterns vary enormously from one car to another. The standards they meet require a headlight to have minimum and maximum illumination values at several different points, but the way an individual headlight illuminates the road between those points may be quite different from the next car over. This may be of interest to you when you're riding at night.

Vehicle positions on the roadway.

Nighttime bicycle/auto collisions show an interesting pattern. Only one quarter of these accidents are the dreaded overtaking from the rear. Three quarters involve motorists coming from the bicyclist's front. That's an interesting pattern, and it makes several points:

  1. Overtaking accidents aren't as big a problem as many people believe, but still worth buying a pair of taillights to avoid;
  2. Accidents from the front dominate, as they do in daytime, because that's where a bicyclist encounters conflict paths at intersections; and
  3. Just because the accident develops in front of the bicyclist doesn't mean the bicyclist figures out how to avoid it. (Also, I suspect that many bicyclists, when they see a car with all its lights on, assume that the driver inside can see the bicyclist just fine. Bad assumption. From outside the car, the lights look blazingly bright, but from inside the car they just illuminate a little patch of pavement in front of the car.)

Because of the limited coverage area of headlight beam patterns, it's easy for a bicyclist on an intersecting path to sneak out of inky blackness into the motorist's path just in time to collide. The accompanying illustrations show this.

By the way, the vehicle position question can take on new significance for overtaking accidents on curvy roads with short sight distances. It may be possible that, under some conditions, the headlights will not bathe the bicyclist's reflector in light right away. (This is an educated guess on my part; I haven't tested it. Still, it's another reason to buy taillights.)

Of the four bicycles in this diagram, only one is even supposed to be in the motorist's headlight beams. The oncoming bicyclist and the bicyclists on side streets are illuminate only by stray light from outside the beam pattern--not enough to bet your life on. The bicyclist being overtaken is illuminated, but only if the motorist's lights are operating properly.

Three fourths of nighttime bicycle/motor vehicle collisions are collisions from the bicyclist's front. The other quarter--collisions from the rear--demonstrate that Murphy's Law hovers over reflector performance.

The human propensity to make big mistakes.

Human mistakes are infinite in their variety.

Here's one example: A bicycle spied in a Boston-area college bike rack had had its rear reflector fall off. The owner had put the reflector back on-with a piece of black electrical tape that completely covered the rear-facing facet of the reflector. Presumably, the person who had done this was a reasonably smart student. But his education just hadn't included this article's section on entrance angles.

A more generic, and common, mistake is to not "see" the bicycle, even though the reflector light is cast upon the retina. Because we see with our brains, not our eyes, the brain can and will suppress that spot of light if it can't make sense of it. Or the brain can mis-identify it, thinking it's just a mailbox reflector.

When you drive, your brain has to "tune out" lots of things. Otherwise, the visual input would be too much. When there's faint visual evidence of a cyclist, the motorist's brain won't "see" the cyclist. Rather, the brain will disregard the faint visual evidence. Only after the visual evidence becomes overwhelming does the brain change its opinion and "see" the cyclist consciously.

I want to talk about fog, because I've been doing fog experiments for 15 years. Back in my youth, I put some reflectors on a phone pole outside my house, and I look at the reflectors every time I drive home at night.

On clear nights, they are visible-although not particularly bright-from 0.3 miles away (the farthest you can get with the phone pole in view). On foggy nights, that can be reduced to 100 feet. And remember, I'm an alerted observer. I know to look for them. I'm sure many others on my street don't ever notice them.

Spoke Divider

Spoke Divider

Lights, please.

"Why rely on the other guy to do everything right?" asks one of my favorite experts in this field, Ohio University Professor of Industrial Engineering Helmut Zwahlen. A reflector functions only when a light is shining on it, the driver's eye is directly behind the light, the driver is going slowly if there's bad fog, etc. etc. etc.

For more than 20 years, the bike safety community has had access to data which list the most frequent collision courses of nighttime bicycle/motor vehicle collisions. These "greatest hits" are entirely predictable, at least in hindsight, when you consider the limitations I listed above.

My friends at the CPSC.

Regular readers of this space-both of you-know that CPSC bashing is a favorite pastime at Cycle Sense. For the most part, they deserve it. But I do see things from their side.

The CPSC doesn't think reflectors alone are safe for nighttime riding. I have asked, and been told in no uncertain terms, "We don't think reflectors [alone] are okay for night riding."

So why do they continue to require reflectors? Well, put yourself in their shoes.

You're the CPSC. You're in charge of preventing human tragedy. Every evening when you drive home from work, you see bicycle riders, solely because their bikes have reflectors. So you know the reflectors are doing some good.

You know that simply revoking the reflector requirement, taking reflectors off bikes, will mean that drivers will hit some of these careless bike riders on their nightly sojourns. Adding a more stringent requirement-requiring lights-will encourage unskilled riders to ride at night, and likely cause more accidents that way. Asking the police to vigilantly enforce bike light laws when they have Kalishnakovs to worry about is a bit unrealistic.

Problem is, the presence of those reflectors encourages the assumption that they are okay equipment for night riding. I've had no success in trying to convince the CPSC that they should mandate a hang tag explaining that the reflectors are an insufficient stopgap measure.

State Laws.

Most state laws don't require taillights at night. Should they? I think so, but I admit that there are weaknesses in my reasoning. Sure, there are a lot of overtaking collisions. But the collisions haven't been studied in a manner that would tell us how many of them would be prevented by taillights (versus how many of them are due to drunk drivers or other such overriding dysfunction). I doubt Congress will fund such a study in the near future. ("Hey guys! We need a quarter mil to study how dangerous it is to ride a bike in heavy traffic at night without lights!")

So out of an abundance of caution, use headlights and taillights, and get your state to enact-and publicize-a taillight law.

Copyright © 1999, 2004 John Schubert You may use this article for your personal benefit. You may not reproduce it or send it to other people. The copyright, including all rights for future use or future publication, remains with the author.

Spoke divider

See also: John Allen on Lights and Reflectors

Spoke divider

See also: John Forester on Lights and Reflectors

Spoke Divider

Spoke Divider

Articles by Sheldon Brown and Others

Reports of the demise of this Web site are greatly exaggerated! We at thank Harris Cyclery for its support over the years. Harris Cyclery has closed, but we keep going. Keep visiting the site for new and updated articles, and news about possible new affilations.

Copyright © 1999, 2004 John Schubert

Harris Cyclery Home Page

If you would like to make a link or bookmark to this page, the URL is:

Last Updated: by John Allen