Clinton Street: Still Too Scary

Two weeks ago, Bike Portland ran a story on Portland's Clinton St. bike boulevard. It included a quote from PBOT traffic safety specialist Greg Raisman:
Our safety performance [on bike boulevards] over time has been excellent. 70 percent of our streets are residential. Less than 20 percent of bike and pedestrian crash activity happens there. … There’s comfort factors that are important in how many cars are on the road. But from a pure safety perspective, the big threat is where you’re crossing the busy streets.
Immediately after finishing the article, I hopped on my bike and rode down Clinton to go to dinner with my girlfriend and her coworkers. Just after I passed 43rd Ave. heading toward Chavez, I saw a group of three cyclists heading in the opposite direction with a Subaru following close behind. I kept my eye on the Subaru, because dangerous passing is a daily occurrence on Clinton. Suddenly and without warning, the car swerved into my lane and started heading straight for me. I had to steer hard to the right, almost colliding with a parked car; the Subaru passed within a foot of my handlebar.
I have been bike commuting for thirteen years, and this is by far the scariest thing that has ever happened to me on my bike. I am thankful to have escaped without injury, but I worry even more about other riders. I had two or three seconds at most to react before a collision happened. What if it had been my girlfriend? She's newer to cycling and can't steer as tightly as I can. What if she hit her brakes too hard and lost control of her bike? What if she avoided the head-on but slammed into a parked car at 15 mph? What about an 8 year old or an 80 year old?
As harrowing as this incident was, this isn't the first close call I've had on Clinton. I've lost count of the number of times I've had to dart off the road to avoid a head-on. I ride Clinton between 51st and 21st to get to work, and not a day goes by without a dangerous pass or two in the morning and another couple in the evening. I applaud Raisman for bringing crash data to the table, but in this case the data can't possibly measure the risk properly. There is an epidemic of aggressive, dangerous driving on Clinton. The only reason why it's not showing up in the crash data is because (thank God) it's been near misses so far. Sooner or later, one of those near misses won't be a miss at all. I am quite certain that if a less skilled rider had been on Clinton that night you'd have read about the crash in the Oregonian, not on some insignificant blog.