What's the methodology for measuring the electricity consumption of bitcoin?
Triangulates energy consumption days from hashprice index, CCAF and Blockware which are in broad agreement now.
What's the error range/margin? 300%?
They are all within 10% of eachother once you factor in PUE consistently across all measures. Cambridge energy consumption has improved a lot since they updated their miner mix in Aug last year.
How is the error margin calculated? This precision requires knowing the exact hardware.