While I mostly agree with you on "Good game design doesn't enforce hard limits to try to force players to do something", good game design and balancing in particular entails "pulling levers" on variables within your control, be it existing or new ones. A member limit can be one such lever, and whether you achieve such a limit by strictly enforcing it or by making it almost completely infeasible (e.g. by a nerf on troop strength or by making treasure production negative or w/e if your kingdom > 60 members) doesn't make much of a difference in this case.
Anyway, I don't think the point is that having a member limit would stop metas in their tracks - it obviously wouldn't. It would only function as a small first incentive against metas. If you want to have a meta you need to organize several wings, which is somewhat more work than simply having everyone in one massive group.
Well spoken Ammanurt . Let's not accept the proposed frame of the debate that hard limits is bad game design.. This is untrue and a lie. Just look around you which games have been among the post popular in human civilization and you know how false such a statement is. This is a false premis of the debate and a frame set by people whom are against a max player limit for other reasons they may not want to reveal.
Hearing people say that a hard member limit is a bad game design in Travian-like games clearly contradicts the fact that Travian Legends are built on this rule. Put this in the light of the dropping player base and closing of local servers in Travian Kingdoms, and the large and frequent updates of Travian Legends recently makes the statement of a hard member limit being bad game design even more absurd.