How far is that from here?

July 6, 2008 5:35 PM

How are the distances between stations on the DC Metro website calculated? They do not appear to be tunnel, walking/driving, or even straight-line (great circle) distances.

For example, this page gives "distances" between Metro Center and all other stations in the Metrorail system. It says the distance to Rosslyn is 1.36 miles, but the straight-line distance between the stations is more than 2 miles (according to this site and this site). Several other station-pairs appear to have the wrong distances listed. It seems to be worse for stations that are far apart. Some station-pairs have listed distances that are larger than the straight-line distance, which makes a little more sense. I am inclined to think the Google Maps distances are more or less correct, so how did WMATA calculate the distances on the website?

posted by thrako to travel & transportation (3 comments total)

I'm not sure if we should trust WMATA on this, but instead of looking at the distance from Metro to Rosslyn I put in the distance from
Metro to Macpherson .41 miles
McPherson to Farragut West .38 miles
Farragut to Foggy Bottome .53 miles
Foggy to Rosslyn 1.26 miles
For a total of 2.58 miles

That seems to be more in line with the google, and with my personal memory or the trip. Basically, all I've managed to do is prove that the WMATA site is completely wrong. Hope that helps.
posted by saffry at 8:02 PM on July 6, 2008

If you sort the list by distance, you'll find that the Regular Fare is still monotonically increasing. That is, for station X and station Y, if distance to X is equal to or greater than the distance to Y, the fare to X is equal to or greater than the distance to Y.

I spot checked from one addition station (King St) and that held. So perhaps the distances they give are more like fare distance rather than actual distance. And if they change their fare structure, they have to artificially change the distances to match how they want to charge between two given stations. If anything, this might lead to lower fares between two given stations, I think. Given they don't appear to list distances greater than the actual. But that's a big more complicated to figure out.
posted by skynxnex at 8:14 PM on July 6, 2008

Thanks for the responses so far. Yeah, if you plot fare against the posted distance for all station pairs you get this picture, with only two outliers. So "fare distance" seems reasonable, though I'm not sure why they don't just give you the actual distance.
posted by thrako at 8:40 PM on July 6, 2008