For scientific testing it would need to be conducted in a controlled environment or be done at nearly the same time and place to ensure similar environment conditions. I guess what I'm trying to say is that if you don't do it at a track or on some closed portion of a highway the results will be questioned. You could try to reduce the amount of questioning by showing a sample format of data, assumptions, considerations and details ahead of the actual results.
But even with that being said I don't see how there is going to be material data presented and conclusions derived that are dissimilar from what we've already produced from others. Maybe if you do a full track test with measurements on each corner at how the car handles the turn. But whatever track picked will be questioned so it should be something that tests speed, acceleration and maneuverability. That is what I'm hoping to see from Top Gear next year, but it would be great to see more in depth data.
As an example: Run both cars on the same track separately taking full measurements and 1st person video noting time of day and environment conditions. Do at least 5 runs each, throw out top and bottom times. Then do it again with each car in front of the other taking video from the back car pointed at the front car as well as the front car video pointed at the car behind. This will give perspective on how each car can or cannot follow into and out of each turn and on straightaways. Then mux the feeds, master the audio, post to youtube and wait for the discussion.
P.S. Don't forget about providing endless detail on the tires like useful tread depth remaining and pressures used. I could see some pedantic discussion there and wouldn't want anything to invalidate the results.