It’s been just over a year since Tucson voters resoundingly shot down the use of red-light cameras for traffic enforcement.

The city manager ordered the cameras shut down for citations the day after the election.

Those cameras, the shells of which remain at the eight major intersections, have been the source of many emails and queries to the Road Runner. So, finally, your humble columnist dove into the most basic question about them.

Did they work? In short, when it comes to curbing dangerous intersection driving, data provided by the Tucson Police Department make a strong case they did. However, the cameras’ impact on collisions — especially serious wrecks — is more difficult to discern, at least according to new data obtained by the Road Runner.

The Road Runner requested red-light camera intersection collision data for two nearly year-long periods before and after the vote. Additionally, a TPD officer provided a spreadsheet with speeding and red-light running violation data — not actual police citations — for the two months after the ban and the same period the year before.

(The cameras did continue collecting data after the Nov. 3, 2015, vote, but not for enforcement.)

According to the two-month violation data, overall incidents of red-light running and speeding through the intersections between Nov. 4, 2015, and Jan. 4, 2016, were up 109 percent and 126 percent, respectively, from the same range a year prior. Data recently compiled by the Traffic Safety Coalition, a pro-red-light camera group funded in part by the industry, shows similar patterns in other Arizona communities where voters ended their use.

During the latter two-month period, that meant 25,291 motorists proceeding into the eight intersections after a light turned red and 5,045 people speeding through them at a minimum of 11 mph over the speed limit. Responding to those findings, John Kromko, chairman of the successful Tucson campaign to end the use of red-light cameras, said he suspects TPD’s violation data was manipulated, possibly by shifting definitions of where the intersection begins.

Some jumps at specific intersections and directions of travel were eye-popping. For example, red-light running on westbound Speedway at North Kolb Road skyrocketed 818 percent to 1,497, and speeding shot up 730 percent on southbound Craycroft Road at East Broadway.

However, the collision data suggest a trend that might be counterintuitive to some: Collisions at the intersections didn’t increase after the cameras went dark, and may have actually declined.

Comparing collisions between October and December 2015 (which includes about a month before the cameras were shut off) to the same period a year before, TPD found that “collisions at all intersections either stayed the same or decreased” with the exception of two intersections, where they rose modestly.

The longer-range data requested from TPD suggest something similar.

That set shows collisions at or near the eight intersections between November 2015 and August 2016 fell 8 percent to 307 from 334 the year prior. Only one intersection, West River Road and North Oracle Road, saw an increase, a modest jump of one to 35. Two intersections had exactly the same number of collisions, and the steepest decline was seen at Speedway and Kolb, where the number of incidents fell 25 percent to 38.

There were no fatal wrecks at any of the intersections in either period, according to TPD.

TPD Assistant Chief Ramon Batista described the mixed bag of data, which shows seemingly incompatible increases in dangerous driving behaviors and flat or declining collisions, as “the wildest anomaly.”

However, thanks to TPD’s Lt. Jamie Brady graciously taking time out of her Veterans Day holiday, the Road Runner received even more fine-grained collision data.

That data confirm the overall reduction in collisions, but add a few interesting twists, including that more serious “reportable” crashes increased slightly and less-serious “property damage only” collisions declined by nearly 30 percent.

David Goldenberg, a TSC spokesman, said that without a more granular look at crashes in the eight intersections stemming from red-light running and speeding, “a complete and full understanding of what is happening” is not possible.

* I understand and agree that registration on or use of this site constitutes agreement to its user agreement and privacy policy.

Anything less is “arguably flawed data. It does not tell a complete or accurate story,” he added.

But Kromko was not at all surprised by the result, arguing that in some ways the cameras made intersections more dangerous, by encouraging sudden braking, among other things.

Also, annual collision data for the eight intersections proper — not including areas approaching the intersection — do show a steady decline from 188 in fiscal year 2006 to 57 in FY 2015, which includes the roughly four-year period during which all of the red-light cameras were deployed, according to a previous TPD analysis. Kromko is also suspicious of these numbers. He provided the Road Runner with an independently produced report based on Arizona Department of Transportation collision data that argues the decline may have just mirrored a broader decline in collisions in the area.

Meanwhile, data from other communities that have shut off red-light cameras show a 16 percent rise in fatal red-light running crashes, according to another report from the Insurance Institute for Highway Safety. Kromko again countered that fatal crashes were fairly rare at the eight Tucson intersections to begin with, and have remained so.

So, what to make of this mess of conflicting numbers?

Batista argued that even if overall collisions are declining, the speeding and red-light violation data suggest that the collisions that do occur might be more serious, an argument supported by the follow-up data Brady provided.

Kromko’s view is that red-light cameras were an expensive scam for drivers when they were operating and at the very least did nothing to improve intersection safety.

For his part, Goldenberg said that if collisions are in fact declining in Tucson, that would constitute an exception to what has been seen in a number of other communities that have banned the cameras.

“The data is indisputable,” he said. “Red-light cameras reduce red-light running, reduce crashes, and reduce the injuries and deaths that occur because of red-light running and speeding.”

But what do you think? Check out the poll available in the online version of this story and weigh in.

Contact: mwoodhouse@tucson.com or 573-4235. On Twitter: @murphywoodhouse