It's be wrong to say the science of Ketones and exercise is pseudo science, but there is a lot of incomplete, and thus mis-, information out there. Because these findings and trends are being reported "live", in order to back up fad diets, social media, bro science, or whatever, it is too easy to jump to conclusions.
Diet is complicated. The scientific method, isolating one variable, is really hard to do with diet. Scientists need to narrow down exercise type (HIIT, VO2 max, typical endurance, or ultra-endurance), training type (elite, sub-elite, amateur, recreational, sedentary), metabolic type ("fat-adapted", traditional, western diet, high protein, high carb, etc.) on top of whatever diet intervention to be able to get a sense of what the role of ketone supplementation really means for performance. Studies only measure one factor at a time, but the peanut gallery will make inferences about all characteristics at once. Athletes meanwhile, figure out what is best even though they don't know why. Like East Africans who don't take any fuel during a marathon. How do they get past what the science behind glucose depletion says should happen in the last 10k? They're already metabolically efficient, often a result of less food availability, or the nature of training groups, which happen early in the morning, without breakfast, for easy, hard, or long sessions.