And yes, it does work, because we have data from the year before this change, to the year after to compare against. The "Eligible Registrations for Beneficiaries with Multiple Eligible Registrations" dropped from 47,314 for FY 2025 to 7,828 for FY 2026. Source: https://www.uscis.gov/archive/uscis-announces-strengthened-i...
I believe the parent commenter's argument is that they instead play the game with multiple people. The increased chance is not per person, but achieved by using more people, each with their own chance.
I don't know if they do this, I merely find the argument itself intriguing with the shift in perspective, and that you as the reader has to keep track of the change in context from the individual one level up.