It’s always fascinating to learn how our favorite services are made, and ex-Google engineer Matt Laroche this week revealed a few Google Maps secrets. In particular, how the popular mapping service calculates how long it’ll take to get from one point to another. It’s something I rely on on a daily basis getting to and from work—even though the ETAs are ballpark estimates—so it’s cool to see how Google accounts for the different real world factors, and how they affect your eventual travel time.
“Google Maps ETAs are based on a variety of things, depending on the data available in a particular area,” Laroche said. “These things range from official speed limits and recommended speeds, likely speeds derived from road types, historical average speed data over certain time periods (sometimes just averages, sometimes at particular times of day), actual travel times from pervious users, and real-time traffic information.”
Laroche goes on to explain some of the deeper details companies, not just Google, use to predict actual traffic time, including live traffic and other data sources. But he admits that calculations are just predictions, and there’s no accounting for the unpredictability of traffic.
“Don’t expect the best predictions to be accurate any time soon,” Laroche said. “Calculating ETAs is a future-prediction problem, and traffic, while it follows certain patterns, is inherently unpredictable.” Even if Google has all the precise data from every phone, speed limit, and road condition, it’s impossible for the search giant’s mapping service to predict a crash or other auto hazard.
Perhaps that’s something we can expect in the way, way, way, WAY future? We can only hope.
Calculating ETAs isn’t as cut and dry as one might think, and is reliant on a number of (obvious) different factors. Maybe remember this the next time you scoff at an ETA that is sort of, kind of accurate. This wouldn’t be a problem if everyone had a self-driving car.