Many companies encourage their employees to report any “near misses” they experience while driving. I hate the phrase “near miss”. Doesn't “near miss” mean they almost missed and therefore did hit? For that reason, I'm going to use “near collision” instead of “near miss”. By getting employees to report close calls, the belief is that identification and discussion of these incidents can help to improve driving performance and reduce future collision potential.

Unfortunately, it's been my experience that when someone does open up about a near collision incident (and, in most cases, they won't come forward on their own), it's usually flavored with the employee's perception that “there was nothing I could do about it; it was the other guy's fault”.

This leads me to ponder two questions:

1. Are near-collisions a meaningful indicator of collision potential?
2. If near collisions are a meaningful risk indicator, does it matter whose fault it was? In other words, is it only an at-fault near collision that indicates risk, or is it all near collisions, even those that “weren't my fault”?

Background on the Data

Before I answer those questions, I first need to establish a little background on the source of my information.

At DriveCam, we have the unique opportunity to identify risky driving through the use of our in-cab video technology. When a client vehicle experiences an abrupt force, it triggers the capture of a short video clip that we then review to isolate any risky behaviors and assess risk. Currently, we have more than 18 million clips with associated review data.

To answer the questions above, I'll draw information from a recent DriveCam study, which looked at near collisions that were both avoidable and unavoidable, during a 26-week period in 2009. During that time period, a group of 46,812 drivers from 210 commercial fleets triggered 700,000 events, of which 133,000 were scored as “risky” and 2,565 involved collisions.

Near Collision Avoidable (NCA)

• A driver with 1 or more near collision avoidable (NCA) was significantly more likely to have a collision than a driver without an NCA. For example, a driver who had 4 NCA's had a 4x greater risk of a collision than a driver with none. Furthermore, a driver with 10 NCA's during this 26-week period had a 7-fold increase in having a collision!
• A driver with 1 or more NCA was significantly more likely to have multiple collisions than a driver without an NCA. For example, a driver who had 4 NCA's had a 13X greater risk of multiple collisions than a driver with none, and a driver with 10 NCA's during this 26-week period had a 46 fold increase! (See Chart Below)

Near Collision Unavoidable (NCU)

• A driver with 1 or more near collision unavoidable (NCU) was significantly more likely to have a collision than a driver without an NCU. For example, a driver who had 4 NCU's had a 4x greater risk of a collision than a driver with none. And, a driver with 10 NCU's during this 26-week period had about a 5-fold increase in having a collision.
• A driver with 1 or more (NCU) was significantly more likely to have multiple collisions than a driver without an NCU. For example, a driver who had 4 NCU's had a 6X greater risk of multiple collisions than a driver with none. And, a driver with 10 NCU's during this 26-week period had a 30-fold increase! (See Chart Below)

The bottom line is that identified incidences of Near Collision Avoidable are a very good predictor of future crash risk. Although not quite as dramatic, the data also shows that Near Collision Unavoidable incidents are also a very meaningful predictor of future crash potential.

This should come as no surprise. It's been my experience that the best drivers constantly scan the traffic picture, maintain adequate distance between themselves and others, and operate at reasonable and safe speeds. With these techniques in action, these top drivers simply make early, well-thought-out adjustments when other drivers make mistakes. Their decisions and actions occur while they're farther away from the other driver.

The less skilled driver commonly has less awareness and limited space. Consequently, these drivers are forced to make abrupt, emergency responses when someone else makes a mistake. So, even if the close call “wasn't my fault”, it's probable the techniques the driver employed increased their chances for a “near collision”.