-
I rode them, I didn't do a load deflection test. But steering and bar leverage against rider weight can all put a twist load into a frame and they vary a fair bit in stiffness there. It won't be all the frame, the whole system flexes bars particularly but from a whippy steel bike to a stiff Al frame and somewhere in between I'd say a perceptive rider can feel it. I'd agree that the finer differences would be hard to call. And I'd not be able to feel varying levels of vertical flex in a frame. I think when most people talk about what they think is vertical flex they're feeling twisting flex (or imaging the whole thing via better fit/comfort).
How much in actual units please? To be honest, I don't believe the difference could be detected in a proper blind user test (which admittedly is almost impossible to do).
Edit: I own a notoriously light and flexible Ti road frame, and when ridden in a turbo trainer the BB visibly moves at least 10 mm side to side. But I'm not at all sure I can genuinely detect that through displacement of my feet, and given it's clamped at the dropouts the tyre flex is removed from the problem.