I can't help but think that there is probably a more rational explanation for this.
For example, the events that they've quoted were all "felt" by the machines.....these machines are all powered and connected to the net.....at the time these events were occuring there would have been a massive increase in net activity, there would also have been a corresponding increase in power consumption, as people went online (and turned on TVs/Radios etc) in order to get more information.
I wonder if that's affecting the results somehow
