A couple of years ago I replaced a cartridge (Audio Technica OC9) as it was getting a bit long in the tooth. It had always had anti-bias set matching the down force figure, or slightly less, as seems to be standard practice.
Upon examining the stylus tip under magnification I could see obvious wear and mis-shaping on the left channel side, i.e. left hand side when viewed from the back of the arm and deck (Michell GyroDec/Audio Origami'd RB300 arm). The right hand channel side was fine.
This cartridge had run for approximately 2k hours, give or take a couple of hundred. This surely must indicate that the cartridge was being unfairly pushed outwards during its life. I have run my current cartridge for around 700 > 800 hours with no bias. When placed on a smooth disc it does pull inwards, but seems fine or if not better when tracking in the grooves! The left/right balance is also spot on when listening too. There is also no sign of undue wear on either side of the tip.
Years back I read a review of a high-end turntable and arm combo (Klimo Tafelrunde turntable with Lancelloto arm) in Hi-Fi World. This arm has no bias compensation facility at all. It seems they agree with me? This is a £17k set up so you would hope some research has been done to come to this conclusion !
You need enormous magnification, around x1000, to see the effective contact area of a stylus, rather than the shank. At such magnification depth of field becomes all but zero so little is in focus, and intense illumination is also required. Also, the cartridge has to be held in a manipulator, both to keep it steady and to get the area of interest into some approximate focus. It’s very difficult and requires special equipment: read expensive. Without all this you were likely not seeing true tip wear.
I cannot speak for the Lancelloto arm and what the designer thinks. During cartridge testing however, mistracking occurs early on one channel with no outward bias force applied, and applying an outward force raises the mistracking threshold (i.e. improves tracking) as well resulting in balanced behaviour from the two channels. So evidence suggests bias force is necessary.
