Stop running battery tests at 65% brightness and actually measure the brightness of the screen
Seriously, this has been an issue in reviews for how long, and how long has it still not been resolved? Do you guys even understand basic math?
Even a 6th grader knows that one screen at 65% and another screen at 65% doesn't mean the screens are actually at the same brightness if the screens have different maximum brightnesses. If one screen is brighter than the other (say 600 nits maximum brightness vs 200 nits for another screen), then 65% of the 600 nit screen is much more than 65% of the 200 nit screen, and thus uses a lot more power (and is clearly unfair for the brighter screens). Anandtech recognizes this and sets all their screens to a standard 200 nits for their battery life tests. The Verge doesn't, and sets up the system in a way that punishes devices with brighter screens. Don't we want manufacturers giving us brighter screens?
Seriously guys, go buy a light meter and make battery tests fairer.