Time warp in Las Vegas?

I don't see the 2% figure in the csicop article.

The CSI article says 20 milliseconds (per second), which would be 2%.

But looking back at the original article, it actually says 20 microseconds, which would be 0.002%. So I take back what I said about the suggestion being inadequate to explain a 2% difference. But I still don't know whether the suggestion makes sense, because I don't know how the device is meant to work.

Brett Tingley at MU has an updated article (dated tomorrow? July 16? maybe I'm in a time warp myself...)

‘Time Warp’ Update and Interview with Joshua P. Warren

If there isn't supposed to be any deviation. Between the two measuring ends of the device. What is the point of having the two measurements?

I'd say the problem here is the quantity of data. With a single anomaly, it isn't possible to plot a graph to visually identify a trend, or do any other sort of assessment of a pattern emerging from the data. It's just a dot.