Skip to Main Content
Common Tread

What Virginia Tech learned about how and why we crash our motorcycles

Nov 21, 2016

What do you learn if you pick 100 riders, put five video cameras and data-logging equipment on their motorcycles and record them for a total of 366,667 miles?

Several things, some of which we knew, some surprising. Intersections are dangerous. We either need to pay better attention or work on our braking techniques, because we crash into the back of other vehicles way too often. We’re not good enough at cornering, especially right turns. And we drop our bikes a lot (probably more often than any of us imagined or were willing to admit).

The study was done for the Motorcycle Safety Foundation by the Virginia Tech Transportation Institute. Of course there’s a lot more to it than those findings above, and I’ll get further into the results in a minute. But first, why do we need some men and women in lab coats to tell us why we crashed?

Motorcycle accident statistics infographic
Motorcycle accident statistics. Infographic by RevZilla

Motorcycle crashes: Complex topic, scarce information

The most commonly cited U.S. study of motorcycle crashes is the one known as the Hurt report. Researchers at the University of Southern California, led by Harry Hurt, went to motorcycle crash scenes to determine the causes. Unfortunately, that report came out in 1981, when cell phones were non-existent and a powerful motorcycle made 90 horsepower. Plus, all those crashes studied were in Southern California.

So even though the Hurt report was the best we had, it was short of perfect. Why does that matter? Well, if we don’t have hard evidence on why crashes happen, how can we make the right decisions to prevent them to keep ourselves safer? Or fight bad legislation intended to protect us from ourselves? Or provide better training for new riders?

How Virginia Tech studied motorcycle crashes

The VTTI researchers recruited 100 riders from age 21 to 79 in California, Arizona, Florida and Virginia. They outfitted their motorcycles with video cameras showing the rider’s face and forward, rear, left and right views. GPS and data loggers captured other information, such as brake pressure, acceleration, etc.

This high-tech approach addressed another weakness of the Hurt report. As thorough as the USC team was back in the late 1970s, they had to gather information from crash scene clues and witnesses, including the riders themselves, when possible. In many cases, they found no evidence that riders took any action at all to avoid a crash, though riders often reported they did. The VTTI cameras and data loggers weren’t likely to change their story after the fact.

While 366,667 miles of riding sounds like a lot, this study still falls short of fulfilling the hopes we had a decade ago of a comprehensive national study. The telling statistic is that in the entire study there were 30 crashes and 122 near-crash events. There are far more than 30 ways to crash a motorcycle, so drawing conclusions from that sample size is tricky. The inclusion of near-crashes helps, however. Sometimes those events teach us just as much or more than a crash.

The VTTI team explains its methodology, including efforts to standardize and define terms and procedures. All the details are in a 20-page report you can download from the MSF. But here are some of the things I picked out.

motorcycles in intersection
Beware intersections. Photo by Lance Oliver.

Where we crash

Intersections. No surprise there. VTTI created a system to calculate how much a certain scenario or riding behavior increased the odds of a crash or near-crash. An uncontrolled intersection presents nearly 41 times the risk of no intersection. A parking lot or driveway intersection is more than eight times as risky and an intersection with a signal is almost three times as risky.

A downhill grade increased the risk by a factor of four while an uphill grade doubled it. Riders were nine times as likely to crash or have a near-crash incident on gravel or dirt roads than on paved roads. And riders were twice as likely to have an incident in a righthand turn than on a straight section of road (crossing the center line is considered a near-crash scenario, even if nothing else bad happens).

How we crash

We complain all the time about other people on the road trying to kill us, especially cars pulling into our paths. The VTTI study partially backs that up. Of the 99 crashes and near-crashes involving another vehicle, the three categories of other vehicles crossing the rider’s path add up to 19.

Here’s the surprise, however. What’s the most common scenario? Riders hitting (or nearly hitting) another vehicle from behind. There were 35 of those incidents. Are we really almost twice as likely to plow into a stopped car in front of us as to have someone pull into our path? Or should we write this off as the result of a small sample size?

Maybe there are clues in the risk section. Researchers tried to break down rider behavior in crashes and near-crash incidents into two categories: aggressive riding or rider inattention or lack of skills. The cameras and other data helped determine, for example, if the rider ran the red light because of inattention or aggressive riding.

The study found that aggressive riding increased risk by a factor of 18 while inattention or lack of skill increased it by a factor of nine. Combine the two, and odds of an incident increased by 30.

Now here's one of the less dramatic findings, but an interesting one, just the same. It seems we drop our bikes a lot. Or at least the riders in the study did. More than half the crashes were incidents some riders wouldn't define as a crash — not a dramatic collision but an incident defined as a case where the "vehicle falls coincident with low or no speed (even if in gear)" not caused by another outside factor. Rider inattention or poor execution are to blame. The study finds "These low-speed 'crashes' appear to be relatively typical among everyday riding," but they are incidents that would never be included in a different kind of study of motorcycle crashes. The cameras, however, capture it all, even our mundane but embarrassing moments.

What we can learn

Of course the practical goal for the MSF in funding this study is to find ways to improve its curriculum for teaching new riders and the study ends with some suggestions. For all of us, however, anything that gets us thinking about where we can be better (and therefore safer) riders is worth a little of our time and thought.

Here’s one thing I know I personally need to work on constantly, and I suspect many of you do, too. We need to look further ahead. It applies on the street, on the track, everywhere. One of the other risk factors the VTTI researchers found that I haven’t mentioned yet is that maneuvering to avoid an object, whether a pedestrian, an animal or something lying in the road, increases the risk factor by 12. Combine that with the high number of riders hitting another vehicle from behind and I get the feeling we’re just not paying close enough attention. We’re not keeping our eyes up and looking far down the road, to see the developing situation that is going to cause the driver in front of us to slam on his brakes, or to spot the hunk of exploded truck tire lying in our lane. Those things are taking us by surprise and we’re not giving ourselves enough time to react.

One thing professional riding coaches teach at the track is to keep your eyes up and look farther ahead. That essentially slows down the action, because you have more time to react to what you see if you’re looking further ahead. If you’re looking at what’s right in front of you on the track (or street), you’re looking at the past, not the future. It’s already too late for you to do anything about what’s 20 feet ahead of your front tire.

The VTTI study isn’t the last word on motorcycle crashes, but that’s OK. There should never be a last word, because we should never stop talking and learning about it.