Google Maps: 13 hours, 5 minutes
Waze: 11 hours, 15 minutes
Pretty much the same path. Screenshots taken at the time, and both programs consider the traffic information.
What accounts for such a time estimate difference? Knowing this will help my scheduling.
It's simple.
Google Maps assumes that you are being a good driver and going no higher than the posted speed limit, and slower if necessary based on the speed of traffic.
When you input your destination into Google Maps your original estimate is made based upon posted speed limits, current traffic conditions, and the distance between your starting point and destination.
Source: https://www.verizonconnect.com/resources/article/google-maps-travel-time/
Waze does not take into account the speed limit and only looks at the speed of traffic. U.S. drivers tend to drive 15-20 mph over the speed limit on expressways (source - personal experience), and Waze assumes you will drive the speed of traffic. Note that driving over the speed limit is (obviously) illegal, so break the law at your own peril - it's not Waze's fault if you get a speeding ticket.
Here's an Medium article written by Waze themselves explaining how they compute travel time. Notice that the article does not mention speed limit as a factor.
I have taken long (4-5 hour) trips using Waze. I always drive the speed limit, which is often significantly slower than the speed of traffic, and I usually end up spending about 25% more time on the road than Waze predicts.
Google Maps is "right" (only if no stops or traffic). Waze is wrong.
I've driven that exact route (Seattle to/from Bay Area via I-5) a couple dozen times. Only once did I manage 13 hours, and that was with only light traffic, literally no stops except for gas (2 brief stops), and consistently 5-10 MPH above the posted limit (which is what most of the rest of the cars were doing).
More typical is around 15-16 hours. That gives time for stretch breaks, stops for food, and possible traffic in the Seattle, Portland, and SF metro areas.
I have just noticed that Waze now has an option to personalize ETA based on past drives:
This may explain the difference with Google Maps. It's quite accurate at predicting my ETA in the US.
An additional possible answer is Competition.
When a smartphone user uses Waze and another app, if they find Waze provides them with a marginally faster route, they'll presume that Waze has a better routing algorithm (at least for that route), and thus choose to use Waze for that route. And so Waze gets more usage hours and advertising revenue.
To achieve this marginally faster route, the Waze routing algorithm may use different metrics and may also apply a scaling to the result. Though there is no real way to verify this without taking a look at the source code...
