Fall back…into traffic?

At least we got an extra hour of traffic sleep, right? 

traffic jam 3

The “time change”: an extra hour of sleep, right? But then there’s the inevitable shortening of the days, the returning home in the dark, and, evidently, the traffic.

Like many who work at Google, most Waze employees commute to work in Mountain View from San Francisco. It’s about 40 miles that tends to take anywhere from an hour to an hour and a half. Last week, after arriving to work nearly two hours later, it seemed as if the normally annoying rush hour were longer, and more annoying. And also confusing: are we crazy, or is traffic just terrible?

We’ere definitely not the first to notice this.

But, with all of this data at our disposal, why not do some research?

The question we posed: Does traffic get worse after the time change?

We wanted to figure out if our commute was longer, and we wanted to do it using the data from our community and the way they report traffic jams. Let’s look at the reporting data from the commuting week (Monday – Friday) for the week (before the week) before the time change (not a typo, just wanted to avoid Holiday traffic since Friday was Halloween). We pulled data specifically for Boston, San Francisco, Los Angeles and New York City, all very large commuter cities with significant userbases:

Screen Shot 2014-11-10 at 9.41.34 PM

This is the reporting graph for the commuting week starting the Monday after the time change (November 3rd):

Screen Shot 2014-11-10 at 10.11.25 PM

You can already see the difference right? No? Maybe a map of of the percentage increase per day will help:

Percent change time change

What a case of the Mondays for the end of Daylight Savings Time! Five different metropolitan areas and only a single one barely comes in under 100% increase from the Monday earlier in the month.

As it turns out, San Francisco didn’t even have the worst of it, though there was still a 75% increase from the control week. May not be as bad as DC or Boston, but it’s all relative, right?