It's telling that Tom's argument was:

If the data is based on the spherical coordinate system of Latitude and Longitude, which you admit is based on the idea that the earth is a sphere, then the results are invalid until you can demonstrate that the system and model is correct.

When it's the result which demonstrates the model is correct. Or, more accurately, it gives confidence in that model.

It's geometrically a spherical coordinate system. Any distance recorded is done under the assumption of a sphere.

Correct. The distances are calculated assuming the earth is a sphere.

The times are not calculated using any assumption, they're just times.

Time = Distance / Speed, so

Distance = Time x Speed

So if the speed is constant then Distance and Time should be correlated.

You would expect airlines to have a roughly consistent cruising speed. Note the world "roughly". Not exactly but it should be close enough. And for longer flights the time for ascending and descent shouldn't have a big impact on the overall time.

So the question is whether there is correlation between the calculated distances and the times?

The graph shows there is. Not exactly - you would expect some variation because of slight differences in cruising speed, time taken taking off and landing and so on. It would be highly suspicious if there was an exact correlation.

But it's close enough to see there is some correlation.

And that gives confidence that the distances calculated are accurate.

And that gives confidence that the model used to calculate those distances, a model of the earth as a sphere, is correct.

The very thing that you're attacking - that the distances are calculated using a spherical model - is the exact point being made.