The "overwhelming majority of tables?" I'd like to see a cite for that claim. If you're talking about Jeremy Crawford tweeting that "the majority of characters don't use feats," you are assuming that the only reason a PC wouldn't have a feat is that they are banned at the table, which is utterly wrong.
  • From levels 1-3, feats aren't available at all (except to variant humans).
  • At levels 4 and 8, feats are competing with "+2 to your prime stat." This is a tough call and many players, especially caster players, will prefer to max their prime stat first.
  • At level 12, you get your first opportunity to take a feat after hitting your stat cap.
So, if you want to estimate how many tables allow feats, level 12 is the most informative: That's the level where a PC is very likely to have a feat unless a) the player just doesn't like feats or b) the table doesn't allow them. Per D&D Beyond, 58% of PCs at levels 12-16 have a feat.

There may be a little selection bias here; high-level tables might be more inclined to allow more complex options. However, if that were a significant factor, you would expect to see the percentage continue to climb as you approach the highest levels, and it doesn't - it stays around 58% through level 20. This suggests that selection bias is not a major factor.

Moreover, at the lower tiers, 34% of PCs grab a feat the instant it becomes possible to do so (levels 4-7), and the number is up to 49% by levels 8-11. That is not at all consistent with the idea that most tables - let alone an "overwhelming majority" - ban feats.

ncG1vNJzZmivp6x7prrWqKmlnF6kv6h706GpnpmUqHylecNmqq2ZpKh6uLTArWStoJViwbq8yJyYpWVlmnqxrdGtsGaZk6nCorjLsmSlp5%2BgwG64yKScZ25ma4N0hI6pmKCdXW0%3D