Visualizing Prediction Inaccuracy

February 26th, 2012
mbtaplot, transit
A major goal of mbtaplot is to visualize bus predictions by converting timing information to locations on a map. One thing I've struggled with is how to represent information getting out of date: the longer since we last got a position update the less accurate our estimate is. I think I've got a good solution now, adding a line back from the prediction to the last reported location. The longer the tail, the less sure we are about the position:






The main thing I like about mbtaplot over nextbus or mbtainfo is that because it gives me more information I can second guess the predictions. The goal with this new feature is to intuitively convey more information about the trustworthiness of the information.

(At some point I want to figure out how to get something like this on my phone; the current version uses too much javascript for that.)

Comment via: google plus, facebook, substack

Recent posts on blogs I like:

The One That I Want Another Sister

I want a baby sister or brother. Because I want to play with another kid, and I don't want to play with Josie because she is in preschool and I won't play with her very often. So I want to play with someone very often who is not just two people. …

via Nora Wise's Blog Posts September 5, 2026

Linkpost for September

On September 19 and 20, Sparrow Infrastructure is putting on The Overhang, a conference in Washington D.C.

via Thing of Things September 4, 2026

Ed Zitron's AI prediction track record

I was curious how well the predictions of the most widely cited AI skeptic I've seen (Ed Zitron) have done, so I looked at how his predictions panned out. To disclose my own biases, I've never had a particularly strong pro or anti AI progress posi…

via Posts on September 1, 2026

more     (via openring)