The Motor Trend description is interesting and they definitely account for more than I expected out them, but unless they're not being detailed, there's a number of variables missed that I mentioned above - some of which are only really issues in certain types of comparisons. And unfortunately much of their compensation relies on each vehicle behaving the same to varying conditions and that's simply not going to be the case.
Conceptually it's a neat idea, but there's zero possibility that their temperature and humidity compensation handles the intricacies of a modern engine ECU, especially in a forced induction engine. Each engine (type) is going to behave differently to each changing condition. Thinking they car apply a correction to handle all weather variations is pretty simplistic. In addition they even mention car to car variations - which they don't/can't correct for. For all you know the slower MKZ in the test was simply a lemon or didn't have enough miles on it for the ECU to finish self tuning.
In addition the specific article that you (presumably) referenced indicates that the tires were not the only changes on the vehicles in question. On the MKZ with the Summer Tire Handling Package that car also has the CCD suspension which could weight the numbers in one direction or the other (slower with weight or faster with less weight transfer to the rear/lifting of the front). In addition the newer car has EPAS which would affect skidpad numbers to some degree and definitely has a weight differential. Unfortunately there's no information there (or in their details of testing) about weight normalization. And that makes sense, when comparing different manufacturers cars. But in trying to quantify a single characteristic you need to change just that one characteristic. Get a single car, run it, reset the car back to the same as at the start (gas, engine temp, etc) change the tires and run it again. The same car, not a different model or year or with a performance pack - unless you're trying to talk about variances with/without the performance pack.
You're trying to pull a single piece of information (tire performance) out of a pair of tests with significant differences.
And yes, I absolutely understand that what I'm suggesting is impossibly to attain - that's the point. Trying to normalize the test results will never work to any real degree. The only reliable method of testing is straight up, with a single driver on a day without changing conditions - I suspect when they test multiple vehicles on the same day their numbers among those vehicles are pretty decent. But, if you want to test tires, I said above, get the same car and change only the tires. Look for a tire testing issue from someone - I'd guess they use the same car on the same track and with similar enough track conditions to get repeatable and coherent results. That's what manufactures (car or tire) do when they objectively test/tune tires for a particular car.
I expect to see some variation considering one tire is a summer only tire and the other is all season. I suspect you'll see a mid-year change to a different all-season and/or an option for summer only tires (or did I miss that in the ordering guide? I have no interest in that, so I might have glossed over it).
