We believe that your comparison of US fracking forecasts creates a false dichotomy between modelling results from the US Energy Information Administration (EIA) and the Bureau of Economic Geology at the University of Texas at Austin (Nature 516, 28–30; 2014).

Our integrated team of scientists, engineers and economists at the University of Texas has built rigorous models that incorporate a wide range of input variables and well-constrained outcome scenarios. In our view, the comparison of just one simulation run with a single outlook from the EIA trivializes a complex problem and fails to represent accurately the rigour and uniqueness of what is being accomplished in our four-year study (see go.nature.com/zfverj).

Your graphic 'Battle of the forecasts' is partially attributed to our data. Although we present preliminary results at conferences and make them available on our website, we explained to the author that our work on the Haynesville and Marcellus plays was not yet finished or published, and requested that it should not be used. We therefore question why you should choose to base the main thread of your argument on a comparison to our unfinished work.

Finally, I find your headline 'The fracking fallacy' potentially misleading: in isolation, it reads as a negative comment on the fracking process itself, rather than on forecasts of natural gas production. Production of US oil is currently at a 30-year high, and of natural gas at an all-time high. Hydraulically fractured wells account for more than half and almost half, respectively, of US natural gas and oil production. To imply otherwise does a disservice to your readers.Footnote 1