My thought, too. Either they use a brand new and unused older phone, or they take the old scores from an old test.That's normal behaviour for a brand-new battery, though. As the battery starts to age a bit it will spend less time at 100% after a full charge.
To be a valid comparison, all three phones would need to have brand new batteries.
Or they compare a brand new phone against who knows how used old phones.
They make as much sense as still having 100% after one hour and drawing any conclusions about how battery life is.