Interesting tool, but I don't want to sign up for the GREET tool to get all of the details on how they calculate emissions.
For example, are they taking into account that very little of the gasoline used in US cars comes from oil drilled in the US? The oil we produce is mostly unsuitable for our own refineries. That is why we export light crude to other countries and import medium crude here.
Are they averaging out the total energy expenditure by location? 15-20% of our oil is from Canada and much of that comes through pipelines, so there is limited additional carbon foot print to transport it. However, as much as 50% of our oil does not come from the Western Hemisphere, meaning it is brought over on tanker ships (which have horrible carbon footprints, especially in international waters).
I had trouble finding good data on what oil from where gets turned into gasoline, which would also affect the calculations. If all the Canadian oil we get was turned into heating oil, plastics, heavy fuel oil, diesel, and fuel for the few remaining oil fired power plants in the US, then the lower carbon footprint of Canadian oil would be more beneficial to the EV calculations. If the reverse were true, and Canadian oil were heavily weighted towards gas, then it would benefit the calculations for non-diesel ICE vehicles more. Similar statements could be made about oil from anywhere.
Given that both this group and the one running the GREET project are scientists, it would make sense that they tried to calculate this. But the only scientists likely to have accurate information about what percentage of oil from which countries is made into specific products are the scientists that work for oil companies. So even considering these extra variables, it would either be done by those without a complete picture, or by those with a vested interest in making the numbers show the oil industry in the best light.