Efficiency is hard to measure because there's such a mix of fuels. Hydro is very efficient, but also a very small part of the generation mix in most places. Nuc

Efficiency is hard to measure because there's such a mix of fuels. Hydro is very efficient, but also a very small part of the generation mix in most places. Nuclear is pretty hard to measure - the actual thermodynamic efficiency of extracting electricity from heat is around 33%, but obviously as that heat doesn't come from fossil fuels it's pretty hard to compare. One way you can look at the grid as a whole is by comparing the average carbon intensity of a delivered kWh of energy with the carbon released by burning when generating the same amount of heat energy. There's a good map of different countries [1] We can compare that with a hydrocarbon fuel like natural gas, which has a CO2 emission of 66.7 kgCO₂/GJ [2]. That's 66.7 gCO₂/MJ or 240 gCO₂/kWh Interestingly, quite a few grids have an "efficiency" (including renewables) of greater than 100%. As an example, the UK national grid had a carbon intensity in 2020 of 181 gCO₂/kWh - meaning using a kWh of energy from the grid releases only 75% of the CO2 that getting that energy from burning gas would. Others are less good - eg Queensland in Australia is at 703 gCO₂/kWh, giving an "efficiency" of 34% This of course includes all the renewables which are pushing up the number. I'd expect it to also include the transmission losses. Interestingly for most countries it's higher than I would have guessed - obviously the ~20-30% renewables and nuclear is pulling up the efficiency. So in summary - I was wrong. The grid is often (depending on your region) a better place to get energy from (in CO2 terms anyway), even if you're just burning it with resistive heat. [1]: https://app.electricitymap.org/map https://app.electricitymap.org/map [2]: https://www.volker-quaschning.de/datserv/CO2-spez/index_e.php https://www.volker-quaschning.de/datserv/CO2-spez/index_e.ph...