We need to make the migration feature more robust and intelligent for users and moderators

I don't think there is a topic more discussed on SE than this topic. A basic search for questions about migration lie in the 4,000 questions asked range. To me this indicates that the current way of handling migrations is insufficient and should be addressed and not simply rationalized away. Due to the length and amount of arguments in the community I will simply list a few posts for example, and then a proposed solution. I am posting this not so much for my idea to be used over others but that we can actually make a step forward in this area. I feel like the justifications to date are "its too much work", and instead we should be making at a minimum smalls feature steps to test if any of the ideas proposed to date would be beneficial to the community at large.

There are currently some open feature discussions on SO and SE on this Flagging migration should include more options and This question may belong to <Another site>, consider migrating and Propose Close -> Migration, as well as some that were rejected such as Migrate based on tags and More options when flagging for migration, but I don't think that the answers have been satisfactory given that the question keeps arising.

I think that we could potentially solve this issue by having a few criteria work together in concert. I propose using the current migration stats like SE already does, use the question tags to make more relevant suggestions, use the flaggers criteria such as reputation (in both SE sites) to expand their list options as well as if the flagger has recieved tag badges (arguably subject experts) to make this system more robust and intelligent for the community as a whole.

In many of the posts the argument is that community members of one community will not know the scope of the recieving community; or more generically that experts are few and limited and would not be best able to judge the correct action. I think this is a short sighted argument. This means the only option is to go to a moderator. To me this burdens a moderator both in time and in needing to confirm the destination community. In addition a queue would be adding to the slew of already exisiting cleanup the community endures (although such questions already go to a queue with some other flag applied in some cases such as deletion).

Proposel:

1. We leave the current commonly migrated based list as it is for those in the community below a certain rep level, such as new users.
2. If you have a tag badge (meaning you are an expert in the topic) and/or sufficient rep level in both the current community as well as the (proposed) destination SE community, then your list includes those sites as options for migration in addition to the ones based on migration stats. This would handle the situation of community scope, because the user knows the scope of both sites due to rep level that they have in both communities.
3. Over time we will see stat trends in tag groupings and can make the system wieghted to list the most likely migration at the top of the list based on historical stats. This is different than todays stats which is an after the fact aggregation. A great example of this is Android or Wordpress which have full communities on these topics.
4. After 3 or 5 flags applied(?) to a question by the experts defined in #2, the question gets migrated automatically OR to prevent all out anarchy, it goes into a queue on the recieving site for that community to make the decision to accept or reject the question. Although I would not want another queue, it would arguably remove this decision from the mods and at least allow BOTH communities to benefit in a sense that the FROM community can boot a question and the TO community can benefit and expand its question and knowledge base.

I know this might not cover any technical hurdles with the current system, but once again I think it's important to take one or two of the proposed options and make headway because this is clearly a painful item in the community that isn't going away and just keeps getting rejected based on opinions of difficulty to implement.