The display below clearly shows the connection between outside air temperature and daily energy use when the temperature is somewhere between 40 and 50° F. The summer of 2013 shows a pronounced dip in daily mean temperature, matched by the dip in the daily electric energy use. If you look, you can see spikes in electric energy use line up with spikes in daily mean temperature.
1. Meters sometimes fail; the common reaction of decision-makers to question the validity of data needs to be answered by looking and understanding the nature of the reported values.
2. I created the display in R version 3.0.2 using the ggplot2 and gridExtra packages. You can produce individual plots using ggplot2 and then configure the elements using gridExtra. The black and gold colors are chosen from a color palette with colors distinguishable for almost everyone, including people with common forms of color-blindness. (Reference: “Using a Colorblind-Friendly Palette”, chapter 12, Winston Chang (2012). R Graphics Cookbook, O'Reilly Media. Kindle Edition, Kindle Location 6813.)