In addition, their rolling distance tests include one factor which most standardised rigs didn’t: the losses from hysteresis when uneven surfaces have to lift/return your mass over a bump. I think soft wide tyres would be even more advantageous than most people realise once real losses on real roads are factored in, and compared at speeds most cyclists go at (i.e most people don’t average 30kph). In fact, even using average trip speeds as the comparator would be daft because that would include downhill sections where riders are actually braking, or don’t really care about rolling resistance because the difference between 50 and 55kph downhill makes no practical difference.
Josh Poertner from Silca is also good but has things to sell.
Of course what everyone is saying above is right -- fitness and rider weight matters the most. But it is fun to look at this stuff especially once you have achieved some level of fitness.