Did the report look at how serious the crashes were?
Yes. And from the looks of things, the number of injury-causing accidents increased after camera installation.
There's actually an unusual balance in the before/after study periods (the "before" is 50 months/4 years, 2 months, while the "after" is only 21 months). The "before" showed that there were no serious injuries ("Injury A") at the intersection, though 13 minor/moderate injury crashes ("Injury B/C") caused by rear end crashes, 3 minor injuries caused by sideswipe and angle collisions. If that figure is averaged out and put into the same timeframe as the "after" period (divide 50 by 21, then divide the accident figures by that quotient), that's 1.26 sideswipe/angle crash injuries, and 5.46 rear end crash injuries.
After the cameras installed, in 21 months, there were no injury-causing sideswipe or angle accidents, but there were 12 minor/moderate injury accidents due to rear end collisions (which, going from the averaged "before" figure of 5.46 above, means the rate of injury-inducing rear end collisions went up by 119%, and injury-inducing collisions total went up by about 78%). The number of non-injury sideswipes also went up after camera installation (from 1 in the 50 months prior (0.42 adjusted) to 3 in the 21 months after). Additionally, the number of accidents at nearby intersections on Mission Street (OR-22) also increased. The number of accidents caused by vehicles traveling on the cross streets decreased slightly, but the number of accidents caused by vehicles traveling on Mission (which the camera was installed to prevent) actually went up. The amount of property damage by the accidents also increased--again, adjusting for the difference in before and after periods, it quadrupled ($16,296 over 50 months before ($6844 averaged out for 21 months) to $27,738 in the 21 months after).
To me, all this seems to point to the notion that ODOT's "waffling" here is clearly about money. They're riding the Redflex gravy train (though their incompetence, amusingly enough, ended up with
a bunch of people getting refunds in Sherwood after Oregonian columnist Joseph Rose exposed that
ODOT "accidentally" deleted the signal timing software at a camera intersection).
Increases in waffles is a good thing.
Yes, if one can eat them. ODOT waffles, sadly, are not edible--give them time, and I'm sure they'll produce a 50-page study showing that. (On another note, is it just me, or would hollowed-out red light cameras make awesome birdhouses?

)
Ultimately, there needs to be a major house-cleaning in ODOT (and the OTC--which needs to become an elected body rather than governor-appointed). There are a few decent folks worth saving, but there's a lot of total clowns in there, too.