-->

## Wednesday, September 18, 2013

### Initial Review of the Automatic Link

I purchased a Link to measure my driving habits. One of the things I was most curious about was the cost of my commute. I have two some equal commute options and I wanted to know if there was a significant savings to one or the other. I only have 18 data points so far (as I split my commute twice a week for a workout) but here's the data so far:

• Route 1 (17.78 miles) Cost - \$2.86/one way • Route 2 (18.55 miles) Cost - \$2.75/one way

Weekly Cost Difference: \$1.0833 • Route 1 Time - 27.25 minutes • Route 2 Time - 28.33 minutes Weekly Time Difference: 10.83 minutes Now, this isn't the full story, since Route 1 is somewhat asymetric. It takes ~25.75 minutes to go and 30.25 minutes to come on Route 1, whereas Route 2 is pretty symetric. It's also interesting to see that Route 1 is shorter and more expensive as my average speed is much higher (there are large portions where I'm going >75 MPH). I should also note that I have 12 datapoints for Route 1 and only 6 for Route 2. Definitely not enough data. Saving \$1 a week at the expense of 10 minutes doesn't really seem worth it. However, Route 2 is much more senic and that gives me 10 more minutes of Podcast time.

Here's a histogram of my trip times:

I'll post again as the data becomes larger.

As for the Link itself, it seems to work pretty well (especially for a beta). It'll be nice as features are added, like an API, so I won't have to manually collect these statistics. I don't necessarily see it changing my habits that much - maybe it would for my wife who makes many more elective trips.