Bad Stats in the QER

I often notice people, even educated people, with a poor understanding of numbers and statistics. Our example today is going to come from the second installment of the Quadrennial Energy Review, released by the Department of Energy. One of the QER’s Key Findings was:

Electricity outages disproportionately stem from disruptions on the distribution system (over 90 percent of electric power interruptions), both in terms of the duration and frequency of outages; this is largely due to weather-related events. Damage to the transmission system, while infrequent, can result in more widespread major power outages that affect large numbers of customers with significant economic consequences.”

Over 90% sounds like a lot. But, the frequency of outages is really not that surprising. In 2014, there was450,000 miles of high-voltage lines and 6,000,000 miles of distribution lines in the United States. This means that 92.5% of lines (measured in miles) are in the distribution system. That means that “over 90%” of disruptions is NOT disproportionate. Actually, it sounds like the frequency of power outages stemming from disruptions on the distribution system is just about what you would expect.

The QER also claims that the duration of the outages on the distribution system is more than the duration of outages on the transmission system. While it’s not as simple as looking at the total miles of power lines to verify if that claim makes sense, it should also not be very surprising if you think about it for a little bit.

I used to live in a house that was at the end of a dead-end. We were the second-to-last house on the road. The start of the dead-end was about 4-5 miles from my house, and the road only had about a dozen houses. Behind our house was BLM forest land. In other words, it was quite rural. We had a power outage one time when a tree fell and took out a powerline during a storm. It happened between our house and the last house on the road, and we were the only two houses which lost power. Since there were only two houses affected, we were lower on the priority list for the power company in responding to that storm. If they have two places where they need to send a crew and one of them affects an entire neighborhood while they other only affects two houses, the two people are out of luck. They’ll have to wait a bit longer before they get their power back.

This can help explain why distribution outages last longer than transmission outages. By definition, an outage to a transmission line is going to affect more people than an outage on a distribution line. So it makes sense that responding to the outage on the transmission line is going to have a higher priority than responding to an outage on a distribution line.