Suntrust Bank, Maple and Berry.
At the last Town Council work session, one of the more disturbing exchanges involved traffic counts. The immediate issue was the implausibly high traffic volume that the Town’s consultant had assumed for the current use of the Suntrust Bank. The consultant assumed that bank building generated 381 trips during the peak hour of the evening rush hour.
This sort of thing matters, as I explained in Post #364. If you overstate the existing traffic, then you understate the impact that MAC development will have on traffic.
Councilman Majdi questioned that figure. And eventually they shut him down with the claim that this estimate was Science. You can hear that about 1:18 into my recording (see Post #450).
As a person who has on rare occasion actually done science, that nonsense claim really got under my skin. You can read Post #364 to see how crude the basic ITE ratebook methodology is. In this case, because this is a very large bank building, and has a (two-lane) drive-through window, the consultant took a figure out a rate book, multiplied by the bank’s high total floor area, and came up with that number as the estimate of current traffic.
So, instead, I’m going to do some actual science. I’m going to count the trips there, during evening rush hour, on a day to-be-determined, next week. Part of actually doing science is announcing that ahead of time, so that I can’t discard the information if it doesn’t match my expectations. (This avoids “publication bias”, in which negative results never see the light of day.)
Is one day’s observation perfect? No. Can I guarantee that counting by hand will be error-free? No. And blah blah blah, for those of you who want to trash-talk this ahead of time.
Will this be good enough to make my point? Yep, I’m fairly confident of that. My point being that the ITE ratebook methodology can be wildly off. To me, the only question is whether the consultant’s estimate is off by more than a factor of ten (“order of magnitude) or not.
Why am I so confident about this? Because I can do simple arithmetic. Check the face validity of 381 trips by doing some long division. That works out to a car going in or out of Suntrust every 10 seconds, steadily, for an hour. Have you ever seen that happen? At that bank? Heck, at any Vienna bank? Double heck, at any Vienna business establishment, period? Maybe one of the busier shopping centers, in its entirety. But at a bank? Not a chance. I walk Maple at evening rush hour all the time (I did so yesterday). Ain’t no way that bank does anything like that amount of business.
The point being that not only is the number wildly incorrect, it was not subjected to even the simplest check of face validity. And if you find one like that, it’s a pretty fair bet you’ve got some other zingers in there as well.
(Need another face validity check? Peak traffic on Maple is something like 2400 cars per hour. The Suntrust 381 trips (in and out) equates to about 190 vehicles. So, Suntrust, by itself, accounts for 8% of the traffic on Maple, during rush hour? Nah.)
I think maybe everybody involved here has lost track of what the ITE ratebook methodology actually is. It’s an agreed-upon methodology used to satisfy certain legal requirements. It’s not science.
This number — and others like it — are why Majdi pressed for using actual (“hard”) counts of traffic in any further analysis. Sure, for the future, we have to rely on some sort of projection. But for the traffic we have now, we can actually count that. And, if the 381 is an indication, we clearly should.
And as an extras-for-experts, I’m going to take apart the ill-defined notion of “bypass trips”. Beyond the counts themselves, the subtraction of assumed “bypass trips” is the next-largest potential source of error in that traffic projection.