I think that the LLNL methodology has changed over time. That, rather than "we've become so much more inefficient," seems a more plausible explanation for chang

I think that the LLNL methodology has changed over time. That, rather than "we've become so much more inefficient," seems a more plausible explanation for changes in useful/rejected energy since 1970. I was recently reviewing the recent LLNL charts. It seems impossible that 2011 and 2016 are using the same methodology: https://flowcharts.llnl.gov/content/assets/images/charts/Energy/Energy_2011_United-States.png https://flowcharts.llnl.gov/content/assets/images/charts/Ene... https://flowcharts.llnl.gov/content/assets/images/charts/Energy/Energy_2016_United-States.png https://flowcharts.llnl.gov/content/assets/images/charts/Ene... They both show 97.3 quads of primary energy use, but in 2011 the chart shows 41.7 quads going to energy services while that dropped to 30.8 in 2016. The US use of energy lost over 25% efficiency in just 5 years? While keeping primary energy use totally flat? The only way to make that work, mathematically, is if people are reducing their use of energy services and preferring inefficient ones for their remaining use. I return to a simpler explanation: the chart methodology for identifying rejected energy and energy services has changed, so you can't compare two arbitrary charts. I wish they would keep the original charts around but also produce a time series showing each year's results with the latest methodology. EDIT: the 2015 chart seems to be the one that introduced new methods; compare to 2014. There's a big shift in just one year.