New site analyzes Tucson bicycle crashes

March 2, 2012 • Posted by: Michael McKisson

A new website documenting bicycle crashes in the city of Tucson launched this week and breaks down the most common types of crashes, where they happen most frequently and the time of day that is the most dangerous.

Bikecolli.info was created by computer programmer and Tucson-Pima County Bicycle Advisory Committee member Colin Forbes.

Forbes, who also operates the bikecolli twitter account, got involved with the BAC and tracking crash reports after he read about an incident in which a Tucson cyclist was hit with a bat and the police wouldn’t do anything despite the cyclist having the vehicle’s liscence plate number.

Forbes said the BAC’s response to accusations that the Tucson Police were biased against cyclists was only anecdotal so the BAC decided to start analyzing crash reports to determine if there was a bias by police officers.

Forbes said while looking to see if there was a bias in the reports, he acquired a mountain of data about crashes in Tucson that he started organizing and analyzing.

“I got caught up by the illusion that if maybe I had one more data point, I could unravel it and make sense of it all,” he said.

The site contains crash data for the City of Tucson as far back as 2007, but he said there are some big gaps in 2010 and 2011. The BAC is working with the city and county to get them to release the reports and waive the fees for the reports.

Forbes said the site is in its first phase and he has plans to break the data down even further.

Currently, though, the front page of the site consists of the information Forbes said he would want to communicate to a cyclist he was riding with for 60 seconds.

Here is the info Forbes found most surprising:

Riding the wrong way on the sidewalk is the most dangerous thing a cyclist can do. “If you stop that particular behavior almost 30 percent of the crashes go away,” Forbes said.

Monday has the lowest crash rate of any weekday. Forbes said he thinks it could be because federal holidays fall on Mondays, which would mean there is less traffic on those days. He said that is only a gut feeling and he has no way of backing that up.

Other interesting findings from the analysis:

22.9% of crashes are right-turning vehicle/wrongway cyclist
16.9% of crashes are crosswalk/intersection crashes
13.4% of crashes are left hook crashes
12.6% of crashes are right hook crashes
The worst area in town in terms of bicycle collisions is 6th Avenue and Congress Street.
The worst intersection is Grant Road and Alvernon Road
There are more crashes in the spring and fall
There are fewer crashes on weekends than weekdays
Most crashes occur between 3 and 6 p.m.

Forbes said sifting through all the crash data actually makes him feel safer riding his bike around town because he knows what to watch out for and how to ride in ways that make him safe.

Be sure to check out the website and look for updates to the data.

← Link roundup: March 2

Two mountain bikers injured in time trials near Phoenix →

8 comments

Newest | Oldest

AskTheLawyers

This is a great tool and will be very useful. The data gathered by this website will help cyclists, motorists and city planners understand what causes the majority of bicycle accidents and therefore help to reduce them in the future. Every city should look into a tool such as this.http://www.californiabicyclesafety.com/bicycling-safety/

3wheeler

Collin is the man.

Wrong-way riding is the most common offense I see while riding, too. Mostly by people old enough to know better. If statisics can enlighten those folks of the reality, then it's all good. Old-time habits are hard to break. Too bad the cops don't seem to get a grant to go out and focus on that particular bahavior. Major thanks to Collin for his efforts and contributions to the BAC's work.

Matt Zoll

We've worked with TPD to ask them to emphasize education/enforcement on wrong way riding in the roadway or on sidewalks or unpaved shoulders, on cyclists riding without lights, and on the few cyclists who flagrantly blow through red lights. We also asked TPD to put the least emphasis on cyclists yielding but not stopping at local street stop signs, which account for something less than 2 percent of all cyclist/motorist crashes based on records Collin, Eric Post and others have reviewed (I may be a bit off on that percentage, but it's very low). We also asked TPD to focus on drivers passing unsafely, speeding and running red lights, and right hooks/left hooks among other infractions. In the fall we're shifting more than $24,000 to TPD for them to focus on improving safety for pedestrians and bicyclists at crosswalks and HAWK pedestrian signals, so they'll mainly be going after driver violations. However, they'll also be enforcing wrong-way bike riding at those locations and any other very obvious illegal actions by pedestrians and cyclists that TPD may not be able to ignore while they're doing their enforcement. However, their primary focus will be drivers. Speaking of crosswalks, we asked TPD to review crosswalk law, case law and a prior city attorney opinion on bicyclists using crosswalks, especially on pathways that have roadway crossings (e.g., Aviation pathway) and at the new bike HAWK crossings (e.g., Swan/3rd Street). They've now issued a memo that TPD officers are not to cite cyclists using crosswalks unless they enter the crosswalk too suddenly to allow a driver to stop or yield, or they ride across the crosswalk at a speed that they determine not to be "reasonable and prudent". Also they indicate that cyclists do not have the same legal protections of pedestrians when using a crosswalk so bear that in mind--although you may not get a citation if you're involved in a crash at a crosswalk, it's likely the driver won't either. And if you do use a crosswalk, including on the Aviation path, be extremely careful because as we know there's a lot of bad driving out there. A big thanks to Collin, Eric and the others who've been working so hard on these crash reviews. It's a tremendous help to target the most critical safety concerns and improve conditions for cyclists throughout the region. Matt Zoll Pima County DOT

Red Star

Yes, great work! Is this being attempted elsewhere? It should and seems grant-worthy...intriguing. Are the histograms at the website supported by rigorous statistical analysis (there being eyeball significance and statistical significance)?

Mike

This is a great resource. Interestingly, I was working on a project which involved me gathering bicycle collision information for the year 2004. Unfortunately, my original research intent did not happen for a number of reasons, but I took the information I did have and analyzed the collision information. It nearly parallels what Collin has found. Grant & Alvernon was still the highest collision location and riding on the sidewalk against traffic was the most frequent type of collision. I understand rynsa's concern about looking at the data in a vacuum. If one was to do a similar collision analysis of all intersections in the city (or county for that matter), you would see that the most collisions occur at busy intersections. So, you need to normalize it based on traffic volumes. Since a similar comparison is difficult to do with bike volumes, I based my analysis on a per capita basis. Yes, it was rough, but it did give a comparison between areas. Anyway, great work Collin!

Collin Forbes

@rynsa:disqus the data primarily comes from three sources, and these are in the "About" link at the top of the page. The first and most immediate source is http://transview.org/TIRS/ . I have monitoring processes which watch their feed and collect new incidents as they happen. Their data seems to come from the police department's Computer Aided Dispatch (CAD) system. The second is from the Tucson Police Department's Crime Statistics Database (http://tpdinternet.tucsonaz.gov/Stats/statistics.aspx). I can usually cross-reference the crashes from TPD against Transview to find the time, date and case number. Third, the BAC requests and usually receives crash documentation from the police and sheriff's departments. These come much less frequently, but let me discern the crash types and other information. Despite cross-referencing, there are still sometimes duplicates from the first and second sources, and the police reports also help remove those. Other researchers (particularly the City of Toronto in a 2003 study) have also used hospitals as an additional source for collision information. I haven't explored that, but it is interesting and could help to fill in additional missing pieces.

rynsa

Very, very interesting site. I'm definitely curious to see what comes of the "Fault Assessment" category in the left sidebar, so I'll be checking in as the fields are populated. I do hope, also, that the authors of this resource will consider clearly labeling their sources -- I'm having a hard time locating where they got all the data. Ultimately, however, I do have a sense that these types of statistical databases can have the unintended consequence of obscuring the issues, so I would warn people to look upon the project with a critical eye. That is to say, this kind of technocratic approach to analysis only presents data in a certain, mathematical light, which can actually compromise understanding, and all while inherently presenting itself as a reliable narrator. Basically, people fall in love with numbers and forget to look at the big picture; forest for the trees, etc. But, I am grateful that Tucson has a community of concerned folks out there looking to better inform the public about bicycles and other local transportation matters. My compliments to the entrepreneurial individuals behind the TBCD (Look! I'm already referring to it with the intimacy of initials!). rynsa