Computerized routing differences

I live in Charlton, Massachusetts. One interesting aspect of Charlton is that the Massachusetts Turnpike goes directly through it, and even has a rest area in it, but there’s no actual access to get on or off the pike in Charlton. (That’s the mixed blessing of limited-access highways after all, that there are few intersections to slow things down, but that means there’s never an intersection quite where you want one.)

So, for pretty much all of my driving life (taking on the habits of people driving me before I could drive, very likely), if I wanted to go westbound, I would take Route 20 West to get on the pike in Sturbridge, and if I wanted to go eastbound, I would take Route 20 East to Route 12 North to get on the pike in Auburn. Quite logical, and doesn’t involve any backtracking really.

Up until recently, routing from online mapping services confirmed these paths as being optimal.

Today, I plugged in an address by Boston (for a PTQ that I’ll be judging there on Saturday), and Google Maps suggests that I backtrack to Sturbridge (admittedly, not really that far) to get on the pike to then take eastbound to my destination. I’m pretty sure this is a change from its prior behavior.

This made me check other mapping services, and both “Live Search” and Yahoo maps continue to suggest getting on in Auburn.

Interestingly, for the return trip, Google suggests getting off the pike in Auburn.

So, maybe I’m just close enough that it doesn’t really matter. But looking at Google’s route, it just might in fact be quicker on the way there and not on the way back.

1 thought on “Computerized routing differences

  1. When you force Google to not backtrack, it looks like it’s a tiny bit slower thanks to being on route 20 for longer, but really not slower enough that backtracking makes sense.

Comments are closed.