I'll try one more time because I don't want this community to wither away, and I feel the only thing to do is retain the new players and keep new ones coming through. If you would be so kind as to share the source of your retention numbers, that would help, because as you can see in responses above, there's some who think retention really isn't an issue.
We don't know the source of a new player. How did they find the project, what were they looking for? Are they SupCom players from years ago that just wanted to play the game again? Did they just search for "Active RTS Games". Can anyone say who these people are or how they got here? Do we have that data?
So I can only theorize, based on my own experiences with how I came to discover FAF, and conversations with other sub 100 game players during matches, that SOME people got here the same way I did.
YouTube > Gyle.
So let's look at Gyle's videos. Here's the data for his last 40 videos (about 1 year) sorted by number of views, notice a trend?
If what people see is these multi layered battles happening across huge maps with a variety of units with UP TO 16 PLAYERS (as advertised on FAF website), then 1v1 and 2v2 Matchmaking aren't going to cut it. Up that matchmaking to 4v4, and block/remove/take away those maps that undermine what's great about the game and you might have it. Perhaps, the "Custom" lobby IS the problem.
This is the thing about bringing a successful product to market. You can't make it what YOU want. You have to make what THEY want. The "THEY" should be the largest portion of the potential users you can find.
Now, if we want to say "No Thanks. The product is what it is. I'm the one who's doing the work of maintaining it, so it's going to be what I want." That's ok. There's nothing wrong with that. But then putting any energy towards wondering why people are leaving is pointless. Because unless their reasoning lines up with what YOU want, nothing's going to change anyway.
I personally don't think the lack of training, education, community events is the problem with retention. Unless there's data to support that. People want to eat the hamburger they saw advertised the way they want. They don't want to be told they're eating it wrong way.
Anyway, thanks for asking the question to begin with.