OK, I just watched the video, and WHAT FLAWED TEST METHOD!!!
I like TFL Car - they're very entertaining. It would just be nice if they came up with some better, more common-sense approaches to testing than they do. In this instance, one VERY obvious, easy get would be to perform each test several times. Especially for the up-hill test, one slick spot that activates traction control early will KILL the run, which is what seemed to happen with the Dunlops.
Downhill: There's no way to know if you're braking on the exact same "straight"-line patch, so an average over several runs will wipe out a bit of error.
Finally, why not do the same runs with TC off and with X-Mode engaged? Seriously, it would have taken them less than an hour, cumulative, to perform the additional testing, and many of us would appreciate the data point.
Again, I like TFL Car - I think they're very entertaining, but they do NOT have a good grasp of the scientific method. I for sure would NOT send them my company's tire for them to test, since the comparison between each snow tire is inconclusive.